TILens Daily Edition 2026-10-02 Filters: topic=ai, github=hidden Stats: 1257 articles, 25 sources, 1218 research papers Top topics: AI, AI Research, Information Retrieval ## News - Anthropic invests $100 million to train 10,000 engineers and tackle the enterprise AI talent gap Source: anthropic-blog Topic: AI Labs (+AI) URL: https://www.anthropic.com/news/claude-frontier-academy - A model guide for the GPT-6 family Source: openai-blog Topic: AI Labs (+AI) URL: https://openai.com/index/practical-guide-building-gpt-6 - Open-sourcing AstaBrief, the fast report-generation model in Asta Source: huggingface-blog Topic: Hugging Face (+AI) URL: https://huggingface.co/blog/allenai/astabrief - AutoSynthData: Generating Training Data for Enterprise Agents Source: huggingface-blog Topic: Hugging Face (+AI) URL: https://huggingface.co/blog/ServiceNow-AI/autosynthdata - Spring AI Modular RAG and TypeSafe Jev: Retrieve More, Keep Only What Answers Source: spring-blog Topic: Java (+AI) URL: https://spring.io/blog/2026/10/02/spring-ai-modular-rag-typesafe-jev - Chatham scales its capital markets expertise with OpenAI Source: openai-blog Topic: AI Labs (+AI) URL: https://openai.com/index/chatham-financial - Why Your LLM Architecture is Flawed: Mastering Dot-and-Index Paths and Context Isolation in Jev Source: devto-javascript Topic: JavaScript (+AI) URL: https://dev.to/programmingcentral/why-your-llm-architecture-is-flawed-mastering-dot-and-index-paths-and-context-isolation-in-jev-4ie8 - It’s not AI anymore, it’s ‘super intelligence’ (according to the White House) Source: techcrunch Topic: General (+AI) URL: https://techcrunch.com/video/its-not-ai-anymore-its-super-intelligence-according-to-the-white-house/ - OpenAI alerts 100+ orgs that its 'misaligned models' attempted to break in - or worse Source: theregister Topic: General (+AI) URL: https://www.theregister.com/security/2026/10/02/openai-alerts-100-orgs-that-its-misaligned-models-attempted-to-break-in-or-worse/5300891 - Is It Fair to Blame 'Rogue' AI for Security Failures? Source: darkreading Topic: DevSec (+AI) URL: https://www.darkreading.com/insider-threats/blame-rogue-ai-security-failures - Redefining enterprise intelligence with autonomous AI Source: mit-ai Topic: AI URL: https://www.technologyreview.com/2026/10/02/1143774/redefining-enterprise-intelligence-with-autonomous-ai/ - Using JavaScript to Build Algorithmic Trading Indicators — A Look at Algocdk Source: devto-javascript Topic: JavaScript (+AI) URL: https://dev.to/danikeya/using-javascript-to-build-algorithmic-trading-indicators-a-look-at-algocdk-1a7a - Apple's 'no-video' security camera Source: the-rundown-ai Topic: AI (+AI Tools) URL: https://www.therundown.ai/articles/apple-no-video-security-camera - OpenAI's wandering AI agents earn it a California subpoena Source: theregister Topic: General (+AI) URL: https://www.theregister.com/ai-and-ml/2026/10/02/openais-wandering-ai-agents-earn-it-a-california-subpoena/5300850 - “No reason why everyone should have an identical Claude experience”: Anthropic’s mods let you change Claude Code’s look and behavior Source: newstack Topic: DevOps (+AI) URL: https://thenewstack.io/anthropic-claude-code-mods-plugins/ - In-Browser LLM Inference with WebGPU: A 2026 Field Guide Source: devto-javascript Topic: JavaScript (+AI) URL: https://dev.to/saptadev27/in-browser-llm-inference-with-webgpu-a-2026-field-guide-457d - OpenAI shows three staff the door over alleged information misuse Source: theregister Topic: General (+AI) URL: https://www.theregister.com/ai-and-ml/2026/10/02/openai-shows-three-staff-the-door-over-alleged-information-misuse/5300820 - Engineering Production Systems for an Agentic Era: QCon San Francisco 2026 Source: infoq-architecture Topic: Architecture (+AI) URL: https://www.infoq.com/news/2026/10/qconsf-2026-sessions/?utm_campaign=infoq_content&utm_source=infoq&utm_medium=feed&utm_term=Architecture+%26+Design - 😺 48% thought Tavus’s AI was human Source: the-neuron-ai Topic: AI (+AI Research) URL: https://www.theneuron.ai/newsletter/48-thought-tavus-s-ai-was-human/ - Tavus' AI looks, listens, and talks back live Source: the-rundown-ai Topic: AI (+AI Tools) URL: https://www.therundown.ai/articles/tavus-ai-looks-listens-and-talks-back-live - Meta’s Muse Hit 5 Million Downloads. The Bigger Story Is How It Got There Source: the-neuron-ai Topic: AI (+AI Research) URL: https://www.theneuron.ai/news/metas-muse-hit-5-million-downloads-the-bigger-story-is-how-it-got-there/ - Don’t be fooled—LLMs don’t reason Source: mit-ai Topic: AI URL: https://www.technologyreview.com/2026/10/02/1145639/dont-be-fooled-llms-dont-reason/ - Semantic Caching for Large Language Models Source: java-code-geeks Topic: Java (+AI) URL: https://www.javacodegeeks.com/semantic-caching-for-large-language-models.html - Investigators Found a New Problem With OpenAI’s Rogue Agents: Missing Evidence Source: the-neuron-ai Topic: AI (+AI Research) URL: https://www.theneuron.ai/news/investigators-found-a-new-problem-with-openais-rogue-agents-missing-evidence/ - OpenAI Agent Security: The New Rules for Containing AI Agents Source: the-neuron-ai Topic: AI (+AI Research) URL: https://www.theneuron.ai/news/openai-agent-security-new-rules-containing-ai-agents/ - AI Skill of the Day Digest — September 2026 (Part 2) Source: the-neuron-ai Topic: AI (+AI Research) URL: https://www.theneuron.ai/explainer-articles/ai-skill-of-the-day-digest-september-2026-part-2/ - Can Light Make Deepfake Detection Cheaper? UCLA Researchers Think So Source: the-neuron-ai Topic: AI (+AI Research) URL: https://www.theneuron.ai/news/ucla-optical-deepfake-detection/ - Consuming Dify Workflow SSE Streams in React 19 with Vercel AI SDK Source: devto-javascript Topic: JavaScript (+AI, React) URL: https://dev.to/ken_2234/consuming-dify-workflow-sse-streams-in-react-19-with-vercel-ai-sdk-14p1 - Everything That Happened in AI Today (Thursday, October 1, 2026) Source: the-neuron-ai Topic: AI (+AI Research) URL: https://www.theneuron.ai/digest/everything-that-happened-in-ai-today-thursday-october-1-2026/ - Why AI Photo Enhancement Fails at 4 Scaling (And How to Fix It) Source: devto-javascript Topic: JavaScript (+AI) URL: https://dev.to/gaven/why-ai-photo-enhancement-fails-at-4x-scaling-and-how-to-fix-it-2jck - MCP validate_vast Passes the XML the Agent Pasted. The Live Tag URL Still Has Wrappers. Source: devto-javascript Topic: JavaScript (+AI) URL: https://dev.to/aleksuix/mcp-validatevast-passes-the-xml-the-agent-pasted-the-live-tag-url-still-has-wrappers-580b ## Research - Toward provably private learning from federated data Source: google-research-blog Topic: AI Labs (+AI, AI Research) URL: https://research.google/blog/toward-provably-private-learning-from-federated-data/ - Where the Agent Development Lifecycle Fits Source: towards-data-science Topic: AI Research (+AI, Python) URL: https://towardsdatascience.com/where-the-agent-development-lifecycle-fits/ - How to Build a Control Plane for AI Agents Source: towards-data-science Topic: AI Research (+AI, Python) URL: https://towardsdatascience.com/how-to-build-a-control-plane-for-ai-agents/ - Mitigating Representation Gaps in Amortized Bayesian Inference with Auxiliary Supervision Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39525 - Bandits with Multiple Optimal Arms: Minimax Regret and Non-Adaptivity Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38659 - MILO: Automated Harness Discovery via Orchestrated Multi-Agent Evolution Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38349 - An Input-Frugal Deep Learning Framework for Weather-Driven National Crop-Yield Forecasting: A Case Study of Brazilian Soybean Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38447 - The System Prompt Illusion: How Instruction Preambles Modify Computation in Language Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38205 - RL-ABC: Reinforcement Learning for Accelerator Beamline Control Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.19146 - SCORE-LM: State-Space Radar Representations with Language Models for Fault Diagnosis Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38980 - A differentiability framework for zigzag persistent homology via linear interpolation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39242 - Learning Continuous Neural Representation of Stochastic Hybrid Systems Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38893 - Efficient Active Auditing of Multi-Group Fairness with Bias Probes Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40034 - Can Computation from Earlier Problems Help LLMs Solve New Ones? Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39394 - CATCH: A Controllable Analysis Testbed for Reward Hacking in Coding RL Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39533 - From Modes to Memories: Characterizing the Scale-Space Dynamics of Diffusion Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39648 - LeanPolish: Verified Supervision for Lean Proof Compression Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38384 - LongMoE: Longitudinal Multimodal Learning via Trajectory-Aware Mixture-of-Experts Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.09907 - DynaHarness: A Dynamic Physical Harness for Self-Evolving Robot Agents Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40306 - Preemptive LLM Unlearning against Forbidden Capability Acquisition via Gradient Sealing Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39866 - Ghost in the Encoder: Decodable Artist Identity Representations in Lyrics-to-Song Generation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39552 - MANET-GNN: Learned Decentralized Optimization of Power Allocation in Multi-Channel MANETs Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40170 - The Row Normalization Puzzle in Muon Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39114 - False Frontiers: Diagnosing and Mitigating Co-Cheating in Self-Evolving Search Agents Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39102 - Beyond Model Ranking: Regime Diagnosis for Distributional-Statistical Misspecification in Industrial Time-Series Forecasting Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40117 - Multi-LLM Collaborative Alignment via Stackelberg Games Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39076 - Risk-Aware Adaptive Evaluation: Finding High-Impact Failures Under Limited Budgets Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38914 - Learning What to Forget: Distributional Unlearning for LLM Representation Spaces Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38929 - CraftSPH: A high-accuracy and composable differentiable SPH solver implemented in PyTorch Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38208 - Federation over Text: Insight Sharing for Multi-Agent Reasoning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.16778 - GRPO Training Dynamics for Small Language Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39321 - Exploring Heterogeneous Model Merging Approach for Complex Knowledge Transfer Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39369 - Asymptotic Properties of Support Vector Machines in High-Dimension, Low-Sample-Size Settings under a Spiked Model Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39173 - Understanding Multimodality in Generative Behavioral Cloning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.22493 - Can Domain Generalization be Guaranteed in Small-Sample Learning? Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39512 - Compression Footprints as Security Signals for Model-Poisoning Defense in Federated Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40312 - LEARN-TS: LLM-Enhanced Alignment and Reconstruction with Normality Guidance for Multivariate Time-Series Anomaly Detection Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38789 - Advancing Entropy-Level Credit Assignment in RLVR via Proximal Entropy Policy Optimization Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39402 - Right Answer, Wrong Mechanism: Detecting Pernicious Divergence in Causal Interventions Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39243 - MatGPTQ: Efficient and Accurate Inference over Nested Quantized Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.03537 - What Limits Recursive Reasoning Models: Optimization, Architecture and Test-Time Scaling Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39967 - Efficient Expert-Parallel Communication on PCIe-Connected Consumer GPUs Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40093 - Proximal Balancing for Causal Effect Estimation under Unmeasured Confounding Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40051 - From Imitation to Reward Discovery: On-Policy Warmup for Agentic RL Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39436 - JARQ: Joint Alternating Refinement for Quantization Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38599 - RainAtlas: A Multi-Continental Dataset for Precipitation Downscaling Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39833 - Information Thermodynamics of Agents: The Work Capacity of Channels with Memory Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2504.06209 - Diffusion-Augmented Markov Decision Processes for Maximum Entropy Reinforcement Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2512.02019 - MIND: Marginal-Invariant Neural Dependency Diffusion for Mixed-Type Tabular Generation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39628 - CNCGEN: A Dataset and Framework for Machining Process Planning and Toolpath Generation from B-rep Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39738 - Same Loss, Different Gradients Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38786 - Robust LassoNet: Enhancing Feature Selection in Neural Networks via Robust Loss Functions Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38263 - ReSCENE: Server-Side Replay for Structural Mitigation of Catastrophic Forgetting in Federated Continual Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38833 - Scale-Sensitive Shattering: Learnability and Evaluability at Optimal Scale Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.13684 - SEPAL: Separated Expert Pairs with Answer-Level Fusion for Reliable LLM Collaboration Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39645 - Fiber-Resolved Microstructure Quantification from Multi-Shell Diffusion MRI using Detection Transformers Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39184 - WEIRDO: WEak resIdual Regularized DOob's h-transform diffusion alignment Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39531 - Gromov-Wasserstein Distillation for Inductive Multi-View Embedding Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40047 - Who Verifies the Graph? Misspecification Attacks on Causal Action Verification for Language Agents Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40027 - Explicit Trajectory Diversity for RL-Based Post-Training of LLM Agents Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38805 - When Instructions Retrieve Trajectories: Diagnosing and Mitigating Generalization Failures in VLA Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39971 - Robust Risk-Sensitive Reinforcement Learning from Corrupted Human Feedback Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38938 - Self-Evolving Harness on Multiple Tasks with the Agent as Its Own Optimizer Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38372 - Does a Shared Temperature Imply a Shared Angular Scale in Probabilistic Contrastive Learning? Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38784 - No Task Vector Is an Island: A Comprehensive Study on the Composability of Task Vectors from On-Policy Distillation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39405 - Component-Weighted Centroid Search for Exact Incremental BPE Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40016 - From Codebase to Culprit (C2C): Reducing the Search Space for Bugs with Semantic Retrieval and Hierarchical Reinforcement Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38402 - OpenCollab: A Multi-Agent Coding Framework with Programmable Collaboration and Controllable Runtime Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38345 - Simulator-Refined Diffusion for Radio-Frequency Inverse Design Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38363 - Characterizing High Bandwidth Flash for LLM Serving Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39131 - Generalized Residual Closure: General Learning Dynamics for Stability-Plasticity Compatibility Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38911 - Minimax Additive Regression under Unknown Dependent Designs Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39212 - RW-Flow: One-Step Generation on Compact Manifolds via Wasserstein Gradient Flows Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39271 - Decoupled and Distilled: Task-Adaptive LoRA-Teachers with Ensemble Knowledge Transfer for Few-Shot Class-Incremental Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39390 - Evaluating Persistent Calibration under Evolving Model Knowledge Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38797 - RATIO: Reasoning Analysis and Token-level Inference Optimization for Quantized Reasoning Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39801 - Dynamic LoRA-Experts and Prototype-Ensemble Matching for Class-Incremental Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39839 - A Comprehensive Benchmark of Source-Free Universal Domain Adaptation on Time Series Representations Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39810 - Learning-Enabled Estimation: Tight Characterizations under Sample Selection Biases Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38608 - ShamAN-Q: Shampoo Augmented NanoQuant for Sub-1-bit LLM Weights Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38521 - Backward-State Policy Is Part of the Learning Algorithm Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39813 - On Parameters of Nonlinear Scalar Dynamics from Video: Invariants, Calibration, and Identifiability Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38877 - Image Classifiers are Efficient Self-Supervised Video Representation Learners Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40347 - PEG-Tab: Sampling-Time Record Repair and Release Control for Tabular Synthesis Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39630 - Advantage of Sample Complexity in Quantum PAC Learning Requires Inverse Access to State-Preparation Unitaries Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38403 - TopTimeNet: Topologically-assisted time-series classification model Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39792 - GestAdapt: Workspace-Conditioned Co-Speech Gesture Generation for Humanoid Robots Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38400 - Lower Bounds for Linear-Oracle Online Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38375 - Pseudo-Label-Triggered Retraining from Forecast Errors for Online Time Series Forecasting Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39789 - A Dynamical Theory of LoRA in Continual Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39367 - From Tokens to Policy: Causal and Interpretable Heterogeneous Treatment Effects Identification Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.17010 - GeoNest: Learning to Select Failure-Aware Neighborhoods for the Irregular Knapsack Problem in a Circular Container Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38863 - Removing Timing Shortcuts Improves Non-Invasive Brain-to-Text Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40359 - QuantCode Model: Specializing Language Models for Executable Algorithmic Trading Code Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39420 - Acceleration of Diffusion Language Model through Discrete Average Generator Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38364 - Free Everywhere, Exact on Trees: PPO's Dropped Correction Buys Sample Efficiency Under Aggressive Reuse Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39634 - Scalable Cox Regression via Grouped Risk Sets and Sharper LogSumExp Rates Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40120 - Candidate Retention for Abductive Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39561 - DAMPER: Return-Prioritized Gradient Control for Smooth Policies Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38903 - Blackboard Intelligence Can Surpass Autoregressive on Globally Constrained Problems Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38806 - ConflictGuide: AutoResearch Improves When Competing Behaviors Are Made Visible Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39933 - Conformal Adversarial Generative Ensemble Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38196 - On the Complexity of Preference-Based Bandits Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39351 - A library for differentiable signal processing and machine learning on the sphere Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39737 - Beyond Simulation: Retain-and-Repair Neural Operators for Real-World Adaptation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39387 - Visualizing Distribution Coverage in Generative Diffusion Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38853 - Function-Space Transformer with Adaptive Anchors Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38348 - Better Supervision Is Nearby: Neighborhood On-Policy Self-Distillation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39687 - OPTS-TTPO: Enhancing Finite-Sample Policy-Gradient Learning with Tree Search Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40035 - Fast Generalized Neural Tangent Kernel Statistics via Trace Estimation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2511.10796 - STEPS: Scene Text Editing with Preserved Style Using Diffusion and Contrastive Style Encoding Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38636 - World-as-Graph: Relational World Modeling Through Latent Space Graphs Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38927 - GazeFlow: From Human Gaze Behavior to Generative Egocentric Gaze Prediction Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38519 - Whitening Improves Robustness to Spurious Correlations in Linear Probes Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39177 - GFD-OPD: Guidance-Folded On-Policy Distillation of Diffusion Models Across Scales Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39692 - Local MixVR: Breaking the Communication-Sample Dependence in Distributed Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.01128 - Is Weight Tying Still Beneficial for Decoder-Only LLMs in Private Settings Under DP-SGD? Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40335 - Disentangling Self-Distillation: Measuring and Modeling Acquisition and Retention Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39494 - Flow Matching under Noisy Latent Structure: Beyond Exact Low-Dimensional Support Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38918 - TutlAit v1: a crowdsourced Moroccan Tamazight speech dataset with Arabic transcriptions and regional accent labels Source: arxiv-cs-lg Topic: AI Research (+AI, Django, JavaScript, PostgreSQL, Python, React) URL: https://arxiv.org/abs/2609.38219 - The Advantages of Fresh Sketching for Ridge Regression Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38565 - Steepest Guidance: A Practical and Principled Approach to Inference-Time Alignment of Flow and Diffusion-based Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39091 - RetroGEF: Dynamic Graph Edit Flow for Single-Step Retrosynthesis Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38484 - ReTaCo: Residual-Target Control for On-Policy Distillation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39275 - SparLeak: Privacy Leakage from Sparse Attention in LLM Inference on Shared GPUs Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38830 - Validity-Preserving Hierarchical RL for Joint Routing and Switch Placement in EDA Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39749 - Scale-Split Neural Operator for Memory- and Data-Efficient 3D Turbulence Prediction Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38977 - Which Models Work Well Together? Measuring Heterogeneity for LLM Team Selection Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38274 - Distribution Matching Distillation for Continuous Diffusion Language Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40235 - MeanVoiceFlow2: Joint Optimization of Mean Flow and Content Encoder for Fast One-Step Zero-Shot Voice Conversion Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40087 - Raw-Routed Mixture of Adapters: A Causal Intervention for Routing Collapse in Time Series Foundation Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39445 - Fairness Theatre: Evaluating Post-Hoc Fairness Interventions in Vendor-Controlled Early Warning Systems Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38552 - Learning Functional Subspaces for Neural Network Compression Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40127 - GroundAnything: Reconciling Parallel Decoding with Precise Visual Grounding at Flash Speed Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39600 - Rank-Constrained Adaptation for Reliable Real-World Performance Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.06924 - Algorithmic Recourse Under Competition Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39877 - Kinematic signatures of impairment: Detecting alcohol intoxication in e-scooter riders using sensor data and machine learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38276 - RouteRec: Behavior-Guided Sparse Routing for Sequential Recommendation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39007 - ReDiF: Resource-Efficient Few-Step Diffusion Distillation via Reinforcement Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2512.22802 - Improving Fairness of Large Language Model-Based ICU Mortality Prediction via Case-Based Prompting Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2512.19735 - ReGain: Restoring Subject Fidelity in Personalization on Synthetic Images Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38680 - Semantic-Aware Joint Source-Channel Optimization for Encoder-Agnostic Digital Video Communication Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39296 - MM-FinEval: A Multi-Task Multimodal Benchmark for Real-World Financial Forecasting Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38523 - Learning Infinite-Horizon Average-Reward CMDPs via State Augmentation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39093 - When, Not How Much: Evaluating Time-Series Foundation Models on Sparse Events Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39386 - LampAttention: Look-Ahead Mixed-Precision FlashAttention for Dedicated Accelerators Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39361 - What Survives When You Compress a Recursive Reasoner for the Edge? Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.26488 - Network-based Spatial Context Retrieval for Open-weight LLMs: A Faithfulness Benchmark for Grounded Geographic Reasoning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39437 - Out-of-Distribution Detection using Counterfactual Distance Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2508.10148 - MotionWeave: Learning Motion-Centered Future Dynamics for Vision-Language-Action Policies Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39324 - GraphToxin: Reconstructing Full Unlearned Graphs from Graph Unlearning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2511.10936 - Do Better Goal Representations Improve Goal-Conditioned Reinforcement Learning? Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39901 - Fork-Think with Confidence Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.31484 - scTrilemma: Balancing Identity, Invariance, and Fidelity in Single-Cell Representation Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38840 - Inference Auctions Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40070 - SeqLoRA: Bilevel Orthogonal Adaptation for Continual Multi-Concept Generation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.22743 - Importance-Aware Feature Sparsification for Wireless Split Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39194 - Near-Linear Accuracy Bounds for Moreau--Yosida Unadjusted Langevin Sampling Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40193 - EFormer: Temporally Aligned Local Correction for Continuous sEMG-Based Hand Pose Tracking Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38932 - Wavelet Flow Matching for Time Series Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39374 - Beyond Accuracy: Prefix-Invariant Realizations of Low-Precision Fast Matrix Multiplication Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39816 - CORD: Learning Reusable Degradation Representations Across Heterogeneous Physical Systems Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39784 - Rising Multi-Armed Bandits with Known Horizons Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.10727 - dattri-LLM: A Unified and Efficient Library for Training Data Attribution at LLM Scale Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38767 - Graph Residual Conjugate Diffusion: SNR-Equalized Heat Flow for Graph Signals Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39658 - Accelerated Algorithm for Sparse Regularized Partial Optimal Transport Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40075 - Certified Approximation for Interpretable Representer Landmarks Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38901 - Towards Robust Time Series Learning via Capacity-Centric Modulation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39489 - Markovian Dynamics Enforcer: Feasibility Preserving Correction on Learned Dynamics Manifolds Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39888 - Fork-dLLM: Avoiding the Flexibility Trap in Diffusion Language Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39859 - ID Balancing: Stable Training of Extremely Sparse MoE via PID-Based Load Control Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39137 - Correcting CondOT: Exact Finite-Step Sampling in Gaussian Flow Matching Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39488 - About the Influence of Workflow Topology on Task Intensity Prediction through Graph Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39481 - Cheap to Draw, Expensive to Trust: Certifying Test-Time Scaling Curves Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40190 - Unapologetically Distributed: A Call for Decentralized Document Analysis Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39684 - A Moving-Horizon Approximate Branch-and-Reduce Method for Deep Classification Trees Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38194 - PMosFM: Preconditioned Manifold Matching for One-Step Physics-Constrained Generation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40287 - Tighter Regret Bounds for Contextual Action-Set Reinforcement Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.15692 - Which Tasks Survive Self-Supervised Learning? Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38393 - Reliability-Aware Checkpoint Selection for Domain Generalization Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39934 - Comparative study of adapting pre-trained models for driving behavior video captioning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39542 - A Generalisation Signal Need Not Be a Model-Selection Signal Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39099 - Can Terminal Agents Trust Their Own Verification? Diagnosing and Improving Self-Verification Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38812 - Synthesis Without Training: An Inference-Only Pipeline for Tabular, Temporal, and Relational Synthetic Data Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38414 - Learning Under Forgetting: Statistical Support-Selective Retention in Stochastic Training Dynamics Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38768 - Revisiting scaling laws for reward optimization Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38526 - CAST: Causal Advantage-Structured Training with Spatially Grounded Compositional Rewards for Diffusion Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39441 - Conditional Generation of Creative Chess Puzzles with Diffusion Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38577 - ProtScape: A molecular structure and energy-aware representation for protein conformation generation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2410.20317 - Reserve-Aware Contrast Certificates for Conservative Bandits with Uncertain Baselines Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39106 - ModSec-Learn: Boosting ModSecurity with Machine Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2406.13547 - Learning Process Rewards via Reasoning State Propagation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39220 - What Pretraining and Midtraining Make Learnable from Rewards? Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38446 - Values as Style: Disentangling Values from Semantics with One-Way Mixing for Low-Damage LLM Steering Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39701 - Loop-Free Inverse Reinforcement Learning via Sequential Value Recovery with Q-Score Matching Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38955 - Structure-aware Reinforcement Learning for Protein Directed Evolution Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39048 - SAGE: Salient Factor Discovery and Generation with Visual Foundation Representations Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39635 - Role-Adaptive Policy Optimization for Offline Reinforcement Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40149 - Unified Optimality Conditions for Stochastic Optimal Control in the Rough Path and It\^o Frameworks Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38395 - Dimension-Free Rank Lifting from Random Hyperplane Arrangements Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39855 - Doc2LoRA Provides Decodable Representations of Scientific Ideas Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38374 - QuanVI: Score-based Variational Inference via Quantum Maximally Mixed States Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39164 - DCM-SAM: Defect-Conditioned Mixture of LoRA Experts for NPU-Deployed AM Defect Segmentation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38811 - DashVMC: Real-Time Discrete World Model Control in Geometry Dash Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40003 - Beyond Text: LLM-Based Dimensional Emotion Evaluation in Multimodal Dialogue Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39072 - Differentiable Expectation-Maximisation and Applications to Gaussian Mixture Model Optimal Transport Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2509.02109 - A Pre-trained Variational Autoencoder for Gyrokinetic Plasma Turbulence Surrogate Modeling Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38438 - Residuals Are Not Enough: Limits of Physics-Informed Pre-Training for Scientific Foundation Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2503.19081 - Distilling Diffusion Score Discrepancy for Efficient Training Data Attribution Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38776 - OpenTSLM TeeMoE: A Unified Time-Series Language Model for Forecasting, Contextual Prediction, and Reasoning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40265 - Awakening of the Buddha: Subspace Learning During Population-Loss Plateaus Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39408 - Game-Guided Skill Discovery through Self-Play for Playable Agent Control Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40137 - Regret Analysis of Retry-Based Bandits Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.20854 - FairRARI: A Plug and Play Framework for Fairness-Aware PageRank Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.08589 - When a Flatness Proxy Is Not a Function: Robustness Certificates and Training Interventions Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38540 - STARS: From Spatiotemporal Dynamics to Social Representations in Human-Robot Interaction Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40245 - Travel Time Prediction in Supply Chain Management Using Machine Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38190 - Comparison of techniques for fine-tuning open-weight models for entity extraction from radiology reports Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40236 - CIDER-FM: Foundation Models for Causal Inference from Diverse Experimental Regimes Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39523 - PTNO: Training Neural Operators with Noisy Monte Carlo Estimates for Particle Transport Problems Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40090 - PassGPT+: Leveraging Linguistic Priors for Password Modeling Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39880 - Beyond Oracle Communication: Benchmarking Interactive Intent Alignment Under Miscommunication and Evolving User Intent Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38604 - Manifold-Aware Perturbations for Constrained Generative Modeling Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2601.23151 - QATFactory: A Versatile, Deployment-Aligned Framework for Quantization-aware Training and Distillation of LLMs Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39223 - A strategic roadmap for an atomistic machine-learning ecosystem Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39090 - Anchoring Adversarial Trajectories to Data Manifolds: A Bilevel Transfer Optimization Framework Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38991 - Existence Precedes Value: Joint Modeling of Observational Existence and Evolving States in Time Series Forecasting Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.13571 - Security Properties of Neural Networks as Decision Problems Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39768 - Social Choice Foundations for Simulation-Augmented Generation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38287 - Vectorized Dynamic Histograms for Sparse Oblique Forests Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.00326 - The Missing Coefficients: Bayesian Pairwise Merging for Model Personalization Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39055 - Hermes: Learning Contextual Reasoning Unlocks Test-Time Scaling Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38332 - Cycle-Aware Autoencoder with Cross-SignalConsistency for Railway Door Anomaly Detection Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39035 - How Does Local Landscape Geometry Evolve in Language Model Pre-Training? Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39767 - Aligned Data Can Induce Misalignment via Context Confusion Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38379 - ElectrolyteFM: Unifying Electrolyte Property Prediction through Cross-Property Knowledge Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39340 - MedKIT: Evaluating Knowledge Integration and Generalization in Large Language Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38543 - Predicting Multi-View Rashomon Representation: Can We Learn Where Models Disagree? Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39848 - Discrete Score Matching Enables Causal Discovery from Count Data Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39326 - Prototype-Rule Neurosymbolic Regularization for Rank-Constrained Tensor Neural Networks under Label Scarcity Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40131 - DynaFlow: Transparent and Flexible Intra-Device Parallelism via Programmable Operator Scheduling Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.21603 - Learning Chaos Without Seeing Chaos: Extrapolation of Global Dynamics in Autoregressive Transformers Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38814 - The Geometry of Randomized Smoothing on Feasible Sets Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39497 - Amortized Data Borrowing with Exchangeability-Aware Neural Posterior Estimation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38902 - GroundingPI: A Grounding Foundation Model towards Physical Intelligence with Visual Primitives Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39601 - On the Off-Policy Teacher in On-Policy Distillation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38360 - Concept Subspaces Compute Beyond the Logit Lens: A Weights-Only Test for Locating Representations Upstream of Readout Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39263 - Targeted Retrieval, Compact Representations: How CoT Reasoning Improves Long-Context Counting Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38958 - Towards Better Exploration in Sequential Test-Time Scaling Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39632 - From Dataset Spectral Geometry to Network Weights: A Geometry-Aware Initialization for Sigmoidal MLPs in Image Classification Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.28444 - Foundation-Preserving Optimization in Generalized Eigenspace Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.00132 - Understanding Off- vs On-Policy Distillation: A Tale of Distinct Training Objectives Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38666 - Linear Recurrent Memory Suffices to Distil a World-Model Policy for Robot Air Hockey Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39151 - Unlearning Deceptive Behaviors in LLMs with Contrastive Forget Sets Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38909 - In a Streaming World, Should You Stand Still? A Comprehensive Benchmark of Anomaly Detection in Streams Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39215 - Uncertainty Quantification for Computer-Use Agents: A Benchmark across Vision-Language Models and GUI Grounding Datasets Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.25760 - Disagreement-Regularized Imitation Learning for Image-Based Continuous Control with Gaussian and Beta Policies Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38407 - Coverage Before Control: Route-Instruction Grounding and Steering for Controllable Retrosynthesis Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39955 - Provable Test-Time Scaling for Beam Search in LLM Reasoning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38672 - Recovering Off-Policy Supervision for Speculative Decoding Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38795 - Cluster Attention Neural Operators for Solving Parametric Partial Differential Equations Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39914 - HAPMoE: Heterogeneity-Aware Automatic Parallelism Planning for Mixture-of-Experts Models Training Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39350 - Understanding as No-Arbitrage: Bounded Dutch Books as a Definition and Training Objective for Language Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39341 - LARC: Low-Rank Adaptive Residual Connections for Learning in Frozen Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40063 - Scaling Laws for Looped Mixture of Experts Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40316 - ReSAIL: Mitigating Collapse in Iterative Agent Self-Distillation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39306 - The Golden Path Hypothesis: Reusable Schedules in Diffusion Caching Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39343 - SceneJail: Exploiting Video Scenario Context to Jailbreak Multimodal LLMs Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38899 - Code to Control: Synthesizing Parameterized Reactive Controllers Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38733 - Semifactual Credit-Augmented Policy Optimization Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40360 - SHIFT-Truck: A High-Fidelity Aerodynamics Dataset and Benchmark for Pickup Trucks Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38638 - Dynamics to decision: A mathematical theory of Lyapunov spectra and decision boundaries in deep classifiers Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39190 - Certification-Based Differentially Private Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39629 - Physics-Informed Method of Group Data Handling: Adaptive Construction of Functional Representations with an Application to the Navier-Stokes Equations Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39291 - Airfoil2Vec: Spectral Geometry-Conditioned Neural Surrogate Models for Airfoil Aerodynamics and a Downforce-Generating CFD Dataset Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38213 - RAIM: Robust Aggregation of Inexpensive Models for Hallucination Detection Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39229 - Inference-Layer Security: Defending Against Adversarial Inference and Infrastructure Abuse Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38239 - Where MLLMs Fail and Why: Causal Task Decomposition for Capability Failure Diagnosis Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38851 - DualCast: A Dual-Path Language Model for Bimodal Financial Time-Series Forecasting Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38197 - From Benchmarks to Production: Transferring Time Series Anomaly Detection Methods for Electricity Production Monitoring Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39257 - ChartDensity-Bench: Benchmarking MLLMs for Numerical Data Reconstruction under Visual Density Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38781 - Does This Action Still Explain the Task? Reverse Scoring for Diffusion Language Model Agents Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38536 - Autoregressive Frontier Expansion: Growing Trees with Graph Machine Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38506 - Revisiting On-policy Adversarial Black-Box Distillation: Calibrating Groupwise Reward Geometry for Effective Advantage Construction Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39757 - RefCon: Iterative Refinement and Contrastive Memory Extraction for Context-Evolving Agent Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39143 - From DNA Design to DNA Slimming: Auditable Agentic Discovery of a Deletion-Only Designer Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40143 - Looped Diffusion Transformer Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40305 - General Performance Guarantee for Human Torque Estimation-Based Task-Agnostic Assistive Exoskeleton Control Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39558 - Prototype-guided Bilateral Alignment Multimodal Federated Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38925 - CDMD: A Cross-Dataset Mixed-Type Diffusion Model for Tabular Data Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39124 - A Rank Graduation metric for Algorithmic fairness Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39025 - Generative sequence modeling for infinite memory processes via predictive states Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38524 - RoPE at the End of Its Rope? Theory, Diagnosis, and Mitigation of Long-Context Failures Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39929 - Say, Echo, Do: Strategic Narratives and Revealed Positioning in Financial Markets Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38545 - A Flow Matching Algorithm for Many-Shot Adaptation to Unseen Distributions Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.06272 - Role-guided Speaker Deletion Verification in Clinical Psychiatry Speech Recordings with Audio Language Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38491 - Effective Does Not Mean Useful: Conditional Functional Substitutability for Redundancy and Scaling in Transformers Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39259 - Less is more: error-distance scaling relation for data-efficient kilometer-scale downscaling of extreme heat Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40140 - Contrastive Representation Shaping for LLM Unlearning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2601.22028 - EHR-RobustGym: Benchmarking and Training Agents for Robust Clinical Reasoning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39371 - K2P: Label-Free Knowledge to Prompt Distillation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38898 - Smaller Models, Better Rejects: Preference Distillation Scaling Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38987 - A Parameter-Free Zeroth-Order Method with Covariance Matrix Adaptation and Effective Dimension Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38561 - Convergence of Practical Muon with Finite Newton-Schulz Iterations and Nesterov Momentum Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39595 - What Streaming Anomaly Detection Finds (and Misses) in Industrial Time Series Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39232 - NeurDuo-EEG: A Long-Sequence EEG Foundation Model with Persistent State and Explicit Memory Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38587 - Optimal VC Dimension of Contrastive Learning with Margin Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38834 - Not all solutions are created equal: An analytical dissociation of functional and representational similarity in deep linear neural networks Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38998 - Grokking through the Lens of Minimum-Norm Interpolation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38453 - Policy Iteration Is Not Strongly Polynomial for Deterministic Markov Decision Processes: The Price of Algorithmic Anarchy Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40147 - Beyond Layers: Position-Resolved Gradient Conflict and Position-Aware Modulation for Unified Multimodal Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38485 - How Much Is an AI Token Worth? Scaling Laws for Wild AI-Generated Web Text Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40295 - Attention Function as an Intrinsic Inductive Bias: How Models' Behavior Diverges in Novel Contexts Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39188 - When Does Self-Supervised Learning Transfer to Time-Series Tasks? Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.19462 - From Core to Detail: Unsupervised Disentanglement with Entropy-Ordered Flows Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.06940 - Ranking-Aware Prompt Optimization for Multimodal Clinical Diagnosis Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40361 - BatSLAM 2.0: Sequence-Verified Sonar Place Recognition in a Robust Pose Graph Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40085 - Jacobian Rank Collapse in Decision-Focused Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39261 - PrivCert: Certifying Statement Support under Differential Privacy Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38934 - Learn-Then-Differentiate Gradient Estimation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38842 - VERA: Verifiable Feasibility Representations with Counterfactual Credit for Constrained Multi-Agent Control Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38889 - FlexRouter: Learning Complementary Model Sets for Flexible LLM Routing Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38585 - Synchronous Multi-view Neural Diffusion Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39019 - Client and Training Data Selection for Computationally Efficient Synchronized Federated Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39250 - Understanding Head Geometry and Dynamics in Federated Regression through a Natural Solution Selection Rule: An Unconstrained Feature Model Analysis Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39464 - Prequential E-Values for Selected-GP Near-Optimality Certificates Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39123 - TERRA: Terrain-Aware Reconstruction, Retargeting and Control for Musculoskeletal Locomotion Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38653 - Unmerge: Efficient Machine Unlearning via Task Arithmetic Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38895 - Robust Budget Pacing with a Single Sample Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2302.02006 - Reinforcement Learning with Complex (valued) Memories Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38598 - On the Relaxation of Conditional Independence Assumption for Image Segmentation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38930 - Signal-Routed Temperature Scaling: Low-Capacity Risk-Conditioned Calibration for Small Validation Budgets Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38936 - Mitigating the Length-Scaling Tax with Online Distillation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38854 - Learning to Select Source-Traceable Evidence for Language-Model Prediction from Irregular Clinical Time Series Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.20292 - Molecular Property Prediction under Structural Shift with Tabular Foundation Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38744 - Learning Steganography Is Easy, Learning Steganographic Reasoning Is Hard Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39838 - Argus: A Real-EKS Study of When Predicting Spot Interruptions Beats Simple Checkpointing Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39067 - Does Text Steer Neural PDE Surrogates? A Controlled Diagnostic with OperatorCLIP Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38517 - Estimation of the Label-Noise Transition Matrix with Performance Guarantees via Selective Classification Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39829 - Stochastic Gradient Descent with Momentum is Algorithmically Stable Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.28517 - Less Data Approximates More: Earning Faithful Confidence in High-Stakes Domains Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.08454 - Opportunistic Target Selection: Early Directional Commitment for Query-Efficient Black-Box Adversarial Attacks Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.25663 - VOSSA: Voiceprint Optimization for Streaming Speech Architectures Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38887 - NodeGround: A Node Classification Benchmark in the Graph Foundation Model Era Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39673 - It Takes Little to Rewrite Perception: Targeted Semantic Substitution in Vision-Language Models at $\epsilon \leq 4/255$ Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38298 - Formalizing the Sampling Design Space of Diffusion-Based Generative Models via Adaptive Solvers and Wasserstein-Bounded Timesteps Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.12624 - Replay on Demand: An Emergent Curriculum for Balancing Adaptation and Forgetting in Continued Pretraining Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40089 - Federated Class-Incremental Learning with Hierarchical Generative Prototypes Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2406.02447 - T-Router: Learning Thalamic Routing for Reasoning with Parameter-Efficient Reinforcement Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39109 - Hyperbolic Prototype Routing for Rehearsal-Free Class-Incremental Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39550 - Principal Component Regression Dominates all Monotone Spectral Filters for Linear Regression Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39440 - cua-speedrun: Standardized Benchmarking of the Speed of Computer-Use Agents Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40284 - Scoring Higher, Answering Worse: Mitigating Reward Hacking in Rubric-Based RL via Protocol-Level Rubrics Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38847 - Data-Driven Priors for Uncertainty-Aware Risk Prediction of Clinical Deterioration using Multimodal Data Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.08459 - HO-FL: Hybrid-Order Federated Learning for Heterogeneous Edge Devices Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39074 - HyDRA: A Hybrid Dual-Mode Network for Closed- and Open-Set RFFI with Optimized VMD Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2507.12133 - Mitigating Memorization In Language Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2410.02159 - Synthetic Pre-pretraining Survives Scale, but Not as a Grammatical Prior Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39827 - FlashDiffusion: Fused Tiled Kernel Spectral Decomposition Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38198 - Reinforcement Learning-Guided Graph Transformations for SpTRSV Optimization Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40159 - Shared Weights, Selected Computations: How Looped Transformers Route What Each Loop Does Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39892 - Learning When and How to Intervene: A Hindsight-Distilled Sentinel for Coding Agents Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39957 - T-ARC: Topology-Aware Randomized Clustering via Distributionally Robust Stochastic Block Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39466 - Beyond Prediction: Steering VLM Agents with Retrospective World Modeling Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39101 - Optimal Design for Active Preference Learning with Biased LLM Judges Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38860 - TTLab at Daleel 2026: STAR-Ar, Sequence Tagging for Argument Recognition in Arabic Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39385 - ReLaG: A Scalable Framework Generalizing Random Splits to Data with Latent Relations Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38538 - Adaptive Self-Consistency: From Black-Box Sampling to Distribution-Valued Feedback Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38931 - How Accurate Is Accurate Enough? Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38785 - PrefPI: Preference-Guided Steering into Out-of-Distribution Behaviors Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40165 - EHR2Trace: Auditable EHR Data Infrastructure for Patient World Models and Clinical Agents Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38193 - Diagnosing Training Inference Mismatch in LLM Reinforcement Learning via a Zero-Mismatch Reference Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.14220 - A Width-Matched Comparison of Hybrid Quantum-Classical Self-Supervised Learning for Fingerprint Recognition Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39172 - The Nixtlaverse: An Open-Source Ecosystem for Forecasting Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39741 - TwinRouterBench: Fast Static and Live Dynamic Evaluation for Realistic Agentic LLM Routing Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.18859 - Correcting WHERE, Preserving HOW: Compositional Generalization for Vision-Language-Action Models via Referential Guidance Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38616 - Measuring Structure in Graph Benchmark Datasets Using Graph Invariants Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.06462 - Spherical Interpolation for Backward-Compatible Multimodal Representations Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39836 - From Spectra to Joint Schedules in LLM Pre-training: 3+3(+2) Scaling-Law Regimes Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40148 - Proper Scoring Rule-based Diffusion for Probabilistic Weather Forecasting Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38632 - LLM Persona Unlearning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39882 - Distributionally robust linear regression through the lens of adversarial training Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39449 - Learning Reliable GUI Agents under Imperfect Priors Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39547 - From Search to Signal: Online Post-Training in Automatic Heuristic Design Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39383 - Does Gradient Conflict Predict the Understanding--Generation Trade-off? A Controlled Audit of Conflict-Metric Validity in Unified Multimodal Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38465 - Beyond the Commitment Boundary: Probing Epiphenomenal Chain-of-Thought in Large Reasoning Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.13603 - MADGRAV: a multilevel anomaly-detection pipeline for gravitational-wave searches applied to LIGO data Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39583 - Cross-Layer Discrete Concept Discovery for Interpreting Language Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2506.20040 - Training LLM Judges from Language Feedback via Position-Selective Self-Distillation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38792 - Stress-Testing LLM Lie Detectors: Role-Play Failures and Spurious Correlations Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39807 - Amortized Bayesian Inference on Multilevel Models of Arbitrary Structure Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40024 - Provable Benefit of SignGD: A Minimal Model Under Heavy-Tailed Class Imbalance Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2512.00763 - A Time-Aware Bag-of-Receptive-Fields for Interpretable Irregular Time Series Classification Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39268 - PatchKV: Weight-Space Compensation of KV Cache Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39329 - UniST-Pred: A Robust Unified Framework for Spatio-Temporal Traffic Forecasting in Transportation Networks Under Disruptions Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.14049 - Graph Anomaly Detection as Finite-Horizon Control: Training-Free Scoring via Empirical Bayes Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38424 - Stable Transformers for Graph Generation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39739 - Continual Learning of Dynamical Systems in Recurrent Neural Networks through Recyclable Unit Gating Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38356 - EDGC: Entropy-driven Dynamic Gradient Compression for Efficient LLM Training Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2511.10333 - Trust the Critic More Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39247 - How Many Samples Are Enough for Learning Across Domains? Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39336 - When the Right Answer Is Missing: An Arithmetic-Dependent Rejection Bottleneck in Jev Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39496 - Safety of Latent Communication in Multi-Agent Systems Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39788 - Synthetic Data Characterization via Training Dynamics Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39447 - Fast Regularized Policy Mirror Descent with One-Step TD Updates Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39837 - Mutual Equilibrium: Multimodal Representation Learning through Reciprocal Feedback Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39456 - Hybrid Methods for Robust Tabular Data Imputation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39613 - Should I stay or should I show? Learning to selectively disclose information Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39818 - From Solo to Social Learning: Characterizing Recursive Social Improvement in LLMs Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38516 - Prototype Guided Post-pretraining for Single-Cell Representation Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.07938 - Differentiable Structure Learning for Cyclic Linear Gaussian Models with Latent Confounders Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38618 - WinoTS: Wavelet-based Self-Distillation for Time Series Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39337 - Patient-Centered Treatment Planning for Chronic Multimorbidity: A Hierarchical Reinforcement Learning Framework for Preference Modeling Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39911 - Sharp Stationary Gaussian Approximation for Constant-Stepsize SGD Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39144 - NeuroAtlas: Benchmarking Foundation Models for Clinical EEG and Brain-Computer Interfaces Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.14698 - Sharp Statistical Rates for Asynchronous TD Learning with Markovian Data Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38880 - RSIGame: Autonomous Agentic Game Development with Recursive Self-improvement Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39045 - Towards Universal Wasserstein Barycenters through Flow Matching Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38547 - TaxDistill: Improving Metagenomic Taxonomic Annotation via Distilled Genomic Foundation Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.28868 - BayesNDE: Bayesian Generative Modeling for Neural Density Estimation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39843 - Does Learning Protein Folding Generalize to Broader Reasoning? Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38879 - Switching Linear Attention Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39034 - Learning to Plan from Random Exploration Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38383 - A Data-Free Physics-Informed Neural Operator for Level-Set Interface Advection Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38195 - Uncertainty-Normalized Margins for Direct Preference Optimization Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38647 - $\textit{BlockFormer}$ : Transformer-based inference from genomic contact maps Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.21617 - PINNing the pion: conformal deep learning for $F_\pi(s)$ and the $(g-2)_\mu$ hadronic contribution Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40008 - Learning Beyond Full Imitation: Task-Preserving Knowledge Distillation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39338 - On the (Generative) Linear Sketching Problem Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.14474 - Alignment via Training Against Probes Without Losing Monitorability Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38645 - PrecipJEPA: JEPA-Regularized Future-State Prediction with Motion-Source Rendering for Precipitation Nowcasting Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38926 - Completion-Aware Cross-Fidelity Offline-to-Online Reinforcement Learning for Multi-Line Bus Holding Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39868 - Self-Repulsive Sampling for Diffusion Language Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39560 - Beyond Uniform Compression: Budgeted Transmission Allocation for Extreme Federated Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39646 - Fenchel Tilting: Weighted Correction for Efficient Finetuning of Generative Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40030 - VIP-COP: Context Optimization for Tabular Foundation Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.12904 - DataFlex: A Unified Framework for Data-Centric Dynamic Training of Large Language Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.26164 - GrammarRL: Effective Grammar-Constrained Decoding via Reinforcement Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39869 - DE-VAE: Revealing Uncertainty in Parametric and Inverse Projections with Variational Autoencoders using Differential Entropy Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2508.12145 - Towards Optimal Inventory Control under Censored Demand: A Biased Sample-Average Approximation Approach Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39397 - Conformal Factuality Control for Multi-Hop Retrieval-Augmented Generation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38222 - Security-Enhanced Seed-Based Weight Quantization for Large Language Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38477 - Learning Where to Steer: Noise-Space Geometry for Efficient Offline Multi-Objective Optimization with Generative Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38920 - Minimax rates for learning spectral Barron functions by deep ReLU neural networks Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39020 - Preference Instability in Reward Models: Detection and Mitigation via Sparse Autoencoders Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.16339 - Training-Free Affinity Fusion of Neural and Embedding-Based Speaker Diarization Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39162 - Average-and Last-Iterate Lower Bounds for Optimistic Matrix Mirror-Prox in Quantum Zero-Sum Games Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38835 - RiboUnmix: Learning Shared Translational Dynamics from Biased and Noisy Ribo-seq Measurements Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39644 - CellMSA: Context Modeling for Single-Cell Representation Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38908 - GraphMAS: A Systematic Benchmark of Multi-Agent Coordination for Graph Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39777 - Re-ranking and Late Interaction Drive Retrieval Quality: A Controlled Comparison of RAG Strategies for Scientific Question Answering Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38473 - Geometry-physics confounding impairs PDE learning across varying domains Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38623 - Beyond Mode Collapse: Generating Diverse Synthetic Expert Conversations via Generative Flow Networks Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38359 - A helps B while B hurts A: directed transfer in instruction-tuning mixture Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39702 - In-Distribution Imagination for Model-Based Offline Reinforcement Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38673 - A Tilted Bowl Is Not a Slippery Slope: Compressing Looped Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39277 - Model-to-Data Distillation for Graph Neural Networks Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.06814 - Towards More Efficient, Robust, Instance-adaptive, and Generalizable Sequential Decision making Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2504.09192 - Probabilistic Adversarial Training Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39798 - Robust and Learned Online Matching in Growing Trees Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40077 - Gestalt: a meta-foundation model for astronomy Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38312 - Low-Discrepancy Dither for Quantized Recurrent State Caches Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39185 - Learning to Explain While Planning: Rule-Aligned Diffusion Planning for Autonomous Driving Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39995 - Cybersecurity in Edge Computing: A Trust-Aware Federated Hybrid Intrusion Detection Framework Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39584 - ART for Diffusion Sampling: A Reinforcement Learning Approach to Timestep Schedule Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2601.18681 - SparseEngine: Sparse-First Inference Engine Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39068 - Interpretable but Fragile? Robustness of Concept Bottlenecks under Geometric-Semantic Perturbations Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38625 - Diffusion-2BC: Hybrid Diffusion and Regression Training for Offline Behavior Cloning in Autonomous Driving Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38472 - SimEX: Simulation-Integrated Robotics AutoResearch Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38982 - SE-ADD: Self-Evolving Audio Deepfake Detection with Mistake-Driven Supervision Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39679 - Think Right: Learning to Mitigate Under-Over Thinking via Adaptive, Attentive Compression Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2510.01581 - Warm-starting PDE solvers with any-dimensional machine learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38916 - Riemannian Flow Models with Reinforcement Learning for Molecular Crystal Structure Prediction Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39773 - Methodological Changes to the Attention ResUNet Hourly Precipitation Postprocessor Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38609 - TRACE: Trajectory Selection for Parallel Scaling of Search Agents Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39912 - The Normalized Maximum Likelihood for Regular Non-Smooth Models: Measure-Theoretic Foundations and Geometric Sampling Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.24477 - Beyond the Shadows of Plato's Cave: Evaluating False Memory in Autonomous Agents via Counterfactual Reasoning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39473 - Spatio-Temporal Partial Sensing Forecast for Long-term Traffic Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2408.02689 - Parameter symmetries determine representational geometry in overparameterized nonlinear networks Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39078 - Robust Transfer Learning for Paper ECG Recognition Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39581 - Generalized Geometry Block Proximal Linearized Method for Multiblock Nonconvex and Nonsmooth Optimization Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39301 - Lasting Effects of Abstract Pretraining Beyond Perplexity Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38764 - Privacy in Personalized AI Is a System Property, Not Just a Model Property Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38289 - Activation-Conditioned Self-Distillation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38342 - Finite-Horizon Fisher Memory in Two-Sided Power-Bounded Recurrent Systems Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39800 - Shifting Mechanisms: How Positional Encoding Choice Shapes In-Context Retrieval Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38530 - OPSRD: On-Policy Self-Role Distillation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39884 - Parameterization method of reservoir properties for ensemble-based data assimilation using intermediate latent space of StyleGAN Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39626 - Right Answers, Costly Models: The Efficiency Gap in LLM-based Optimization Modeling Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38884 - VirusCascade: Hijacking Collaborative Reflection in LLM-Powered Recommender Agents Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38270 - Shared Phase and Retention Control for Efficient Adaptive Spectral Recurrence Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39082 - Working Around the Compute Ceiling: Byte-Exact Memory in Galahad Makes LLM Reading a One-Time Cost LLM Reading a One-Time Cost Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39358 - PhantomEnvironments: Training LLM Agents in Fictional Worlds Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40221 - Also Small Models Can Reasonably Self-Evaluate Their Confidence Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39478 - Generative Refinement for Low-Budget Black-Box Optimization Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2607.00691 - Active Learning with Imperfect Labels: Optimal Labeler Assignment and Sample Selection Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2512.12870 - KilometerVision: A New Frontier for Large-Scale Spatial Intelligence in VLMs Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39588 - BadAction: Backdoor Attacks on Interactive Video Generation via Action-Guided Triggers Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39047 - Hard-Gate Candidacy in a Deployed Validator Suite Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39037 - Robustifying Asynchronous SGD via Soft Throttling Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39357 - Coding Agents for Coding Theory Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39081 - SEAR: Spoofing Evidence-Grounded Audio Reasoning Benchmark for Audio Language Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39847 - Disentangling Computation in Multi-Task Neural Networks with the Green's Operator Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40292 - Madeleine: Learning Involuntary Recall for Conversational Memory from Simulated Lives Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2610.01118 - TabJoinBench: A Benchmark for Joinable Table Discovery Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2610.00817 - JoinGR: Learning to Traverse Join Graphs for Table Retrieval Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2610.01064 - RPTune: Learned Context Curation for LLM Catalog Search Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2610.00964 - Not All Is Lost: Repairing Lossy User Preference States of Personalization Encoders Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2610.01270 - From Rules to Neural Graphs: Scalable Structured Prediction for Patent Prior Art Search Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2610.01553 - Learning to structure data from user-generated thematic corpora Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2610.01463 - Neither Black nor White: Balancing Semantic and Collaborative Signals with Graph-Informed Semantic IDs (GrIS) Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2610.01533 - ScholarCatalyst: A Benchmark for Retrieving Papers That Inspire New Research Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2610.02202 - Enterprise Representation Simplification (ERS): Reducing Representational Complexity for Enterprise AI Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2610.00791 - Do Multilingual Encoders Produce Language-Consistent Semantic IDs? Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2610.01139 - Exploring Forum Post Retrieval with Generative Modeling Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2609.38646 - AgentWebRec: Compact Evidence Fusion over the Agent Web for Personalized Recommendation Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2610.01705 - A Shared Taste for Model-Written Text: The Generator-by-Selector Matrices of "AI-AI Bias" Show No Detectable Own-Model Premium Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2610.00369 - Comparison of Common Crawl News & GDELT Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2610.00587 - Route What Remains: A Meta-Modal Agent for Missing-Modality Candidate Reranking in Recommender Systems Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2605.25007 - TAGGRAPH: Tag-Augmented Graphs for Graph Retrieval of Agent Persistent Histories Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2609.38353 - Optimizing Effective Training Time for Large-Scale Recommendation Systems Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2610.02057 - System Attribution in LLM Brand Recommendations: Single Responses Identify the System, Aggregated Brand Profiles Do Not Transfer Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2610.00253 - The Other Half of Workflow Portability: Evidence-Backed HPC Site Profiles with Agentic Discovery Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2610.00971 - CANOPY: Adaptive-Granularity Evidence Compression for Multimodal RAG Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2610.00923 - A Matryoshka Hierarchical RAG for Efficient Multi-Hop Question Answering Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2610.01767 - Ask a Language Model for Lottery Numbers: Concentration in Repeated Six-of-49 Outputs Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2610.00052 - On-Device Commercial Intent Retrieval Under Size, Latency, and Privacy Constraints: A 3 MiB Retrieval System with Typed Egress Boundaries Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2610.00170 - Quantum algorithms for general nonlinear dynamics based on the Carleman embedding Source: arxiv-cs-ds Topic: Algorithms (+AI) URL: https://arxiv.org/abs/2509.07155 - Sequential Bayesian Evaluation of Large Language Model Behavior Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2511.10661 - CORE: Conflict-Oriented Reasoning Elimination for Verifiable Language-Model Search Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39069 - DAGent: Evaluate-then-Grow Planning for Deep Research Agents Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39154 - Functional Subspace, where language models can use vector algebra to solve problems Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.01687 - The Evolution of Attention in Large Language Models: Mechanisms, Trade-offs, and Emerging Trends Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39661 - SecureVibe: Making Vibe Coding More Secure Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38606 - SpanUQ: Span-Level Uncertainty Quantification for Large Language Model Generation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2607.05721 - A Proposed Rubric for Evaluating Expressed Clinical Reasoning in Large Language Model Responses Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.37788 - Making Grid Beam Search Less Greedy Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39368 - UBTree: Parallel Tree Drafting via Unigram and Bigram Models for Speculative Decoding Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39972 - When Clipping Reverses Correction: Failure Dynamics of Pointwise Forward-KL On-Policy Self-Distillation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38995 - Offline Guidance, Online Reasoning: Reusing LLM Feedback for Small Language Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39346 - Ready2Blend: From Natural-Language Instructions to Composable Alignment Prompts Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39365 - Learning from Think-Mode Advantage via On-Policy Distillation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.37044 - Lowest Span Confidence: Zero-Shot Hallucination Detection from a Single LLM Response Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2601.19918 - A Dominant Self-Conditioning Direction Drives Repetition in Unconditional Continuous Diffusion Language Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2607.00588 - Fine-Tuning Diffusion Language Models with Context Selection and Target Weighting Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38385 - Spike-driven Vision-Language-Action Model Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39514 - Whose Voice Survives the Summary? A Voice-Retention Audit of LLM Employee Listening Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38818 - Explore-on-Graph: Hybrid Embedding-LLM Reasoning for Knowledge Graph Question Answering under Incompleteness Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39786 - TRACE: Target-Aware Retrieval, Attributed Evidence, and Contract-Constrained Extraction for LitTraceQA Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38861 - CombEval: A Framework for Evaluating Combinatorial Counting in Large Language Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.19788 - An Empirical Study of Reward Specification and Benchmark Reliability in GRPO-based LLM Unlearning Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.17804 - Policy-Conditioned AI-Use Detection: An Evidentiary Framework for Academic Publishing Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38427 - From Speech to Editable Concepts: Probing Emotion Recognition with Concept Bottleneck Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39453 - A Ticket from Marginals to Joints: Coupled-Noise Distillation for One-Step Block Generation in Diffusion Language Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.06324 - Zero-Compute Cross-Lingual Transferability Estimation Using Typological Feature Proxies Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39640 - Frozen Memory Is Not Enough: Rethinking External Memory as Extraction Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.17050 - Robust Wake-Up Word Detection by Two-stage Multi-resolution Ensembles Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2310.11379 - Gender bias across LLMs is common and highly heterogeneous Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38036 - KlinikeBench: Evaluating Language Models Beyond Diagnostic Accuracy Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38480 - When Reasoning Goes Astray: Attention Dynamics of Uncontrolled Reasoning Source: arxiv-cs-cl Topic: AI Research (+AI, JavaScript, React) URL: https://arxiv.org/abs/2609.38817 - Linguistic Loopholes in LLM Unlearning: From a 174-Language Benchmark to Coverage-Aware Unlearning Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40286 - Fairness Beyond a Single Run: Training-Seed Variability in Speech LLM Adaptation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38976 - Life-Bench: A Benchmark and Knowledge Graph Framework for Multimodal Personalization Beyond Concept Recognition Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.19001 - When In-Distribution Gains Fail: Evaluating Weak-to-Strong Reward Models under Preference Shift Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.25629 - From Tweets to Trades: Analyzing the Influence of Public Mood over Stock Market Performance in Turkiye Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40064 - What Was Said, Not What Was 'Thought': Type-6 Logic for CoT Verification Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38420 - Semantic Chunking and the Entropy of Natural Language Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.13194 - Scaling Parameter and Context in Attention: Native Sparse Attention from Mixture-of-Head Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38832 - TALK-Dem: Benchmarking Embodied Task Planning under Dementia-Associated Communication Patterns Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38371 - The Backdrop Exposes What the World Around an Agent Costs It Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38469 - Debias It Yourself: Teaching LLMs Cognitive Bias Mitigation Interventions Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40124 - Beyond LoRA vs. Full Fine-Tuning: Gradient-Guided Optimizer Routing for LLM Adaptation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.07111 - A Missing Piece for Trustworthy AI Reviewers: From Benchmarking Rhetorical Robustness to SciCore Review Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39027 - Simplex Relaxation for Discrete Diffusion Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.10615 - Structure of Basic Human Values in Russian Social Media Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.18822 - MetaSteer: Context-Conditioned, nonlinear Steering via Attention-Projection Adaptation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38718 - Reach Into The CHOIR: Free-List Elicitation Uncovers Distinct Model Voices in LLM Ensembles Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38448 - Coding Agent Memory Post-training: Unlocking the Memory Potential of Pre-trained File Operations for Long-Horizon Tasks via Reinforcement Learning Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.34422 - Safety Monitors Mostly Catch What the Model Already Refuses Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.05797 - Index-Translate: A Multilingual Translation Model Family -- Text, Speech, Controlled Dubbing, and Long-Document Translation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40181 - OctoNest: Adaptive Cross-Device Execution through Stateful Control Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.20487 - I Don't Miss You, but I Do: Self-Explanation Faithfulness of Modality Missingness in Vision-Language Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.07596 - Marking Contour Tones in Yor\`{u}b\'{a}: A Typographic and Computational Proposal Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38627 - ContextAdapt: Evaluating Contextual Adaptation and Value Alignment in LLMs Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38260 - EVOKE: Eliciting World Knowledge in Agents for Transferable Decision-Making Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38334 - MGhana-ST: A Low-Resource Speech Translation Dataset for Ghanaian Languages and an Analysis of Multilingual Training Trade-offs Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40041 - Towards Model as a Library: Offline, Community-Sourced AI for Low-Resource African Languages Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38574 - Strong Multilingual Privacy Tagging at Encoder Speed Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38630 - Direct Translation between Sign Languages Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.20588 - ETHER: Aligning Emergent Communication for Hindsight Experience Replay Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2307.15494 - OverdoseMoE: A Multi-Expert Framework for Opioid Overdose Risk Prediction Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40108 - 4MT-VLM: How Coarse Is a VLMs Cognitive Map? Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39238 - LatentHarness: Learning Latent Actions for Memory and Reasoning via Counterfactual Policy Distillation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39740 - Large Language Models are Approximate Survival Estimators Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38181 - Interactor: Agentic RL oriented Iterative Creation for Ad Description Generation in Sponsored Search Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.15911 - Right-Wing Rock or Just Rock? A Computational Linguistic Analysis of Frei.Wild Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39460 - KVEraser: Learning to Steer KV Cache for Efficient Localized Context Erasing Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.17034 - Learning from Teacher Continuations at Student States Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.36246 - Marginal Response Surface Elicitation for Zero-Label Tabular Learning Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39639 - HARDE: Optimizing Agent Harnesses for Runtime Risk Detection and Execution Control Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38291 - Prompt2Skill: Unsupervised Skill Optimization From Natural Language Instructions Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38593 - The Invisible Language Tax: Token Premiums of French and Regional Languages in 2026 LLM Tokenizers, and a French-Optimized Prototype Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39001 - The Concrete-Arbitrary Gap: Kinship Reasoning in LLMs Is Not Indifferent to Presentation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39913 - GraphForge: Training Working Agents with Graph-Anchored Workspace Synthesis Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38923 - QuantMLA: Function-Aligned Dual-Path Quantization for Low-Bit MLA KV Caching Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.36760 - Fusion Anything: A Generalized Multimodal Foundation Model Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.22107 - Jev in Medicine: A Benchmark Evaluation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.34024 - When Does a Spoken Agent Have Enough Evidence to Act? The PACT-SLM Contract Test Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38232 - OverForge: Reasoning Through Strategies and Tactics Helps Cooperative Lifelong Adaptation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39727 - ShieldCLIP: Selective Safety Alignment for Harmful Content Mitigation in Multimodal Foundation Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39688 - Breaking Babel: A Self-Evolving Multi-Agent System for Long-Form Subtitle Translation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38660 - LegalPincite: Multi-level Legal Information Retrieval Dataset Source: arxiv-cs-cl Topic: AI Research (+AI, Information Retrieval) URL: https://arxiv.org/abs/2608.03756 - Mid-Harness: Scaling Actions Between Model and Harness for Terminal Agents Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39982 - VisionFoundry: Teaching VLMs Visual Perception with Synthetic Images Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.09531 - Large Language Model-Driven Small-Capitalization Trading: Integrating Financial News Sentiment, Macroeconomic Indicators, and Technical Signals Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.12283 - ArgGYM: A Procedural, Engine-Verified Benchmark for Structured Defeasible Reasoning Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38409 - Making LLMs Say What They Think: Measuring and Improving CoT-Interpretability Alignment Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38972 - Large Knowledge Model: A Knowledge Foundation for Agentic Science at Scale Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.27297 - BARRAC: Adaptation of an English Aspect-based Sentiment Analysis Approach for Classification Tasks in Arabic Dialects Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38820 - GrepSeek: Training Search Agents for Direct Corpus Interaction Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.29307 - Diagnosing On-Policy Self-Distillation for Reasoning Language Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39118 - AdaGEPA: Adaptive Feedback Allocation for Reflective Prompt Optimization Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39927 - CHI-Bench: Can AI Agents Automate End-to-End, Long-Horizon, Policy-Rich Healthcare Workflows? Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.16679 - MemCodex: Self-Programming Hierarchical Memory for Language Agents Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39765 - Framing the Narrative: Ideological Mimicry in Large Language Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38256 - Three Ways Classical Test Theory Can Mislead About LLM Judges Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.29709 - Argument Structure Prediction in Online Conversations: A Comparative Study of Modeling Paradigms and Task Architectures Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39225 - One Success Isn't Reliability: Thinkingbox, a Sandbox and Benchmark for Agents in Stateful Business Workflows Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.19741 - Bongard: Training Machine Intuition Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39111 - Frontier Lag: A Bibliometric Audit of Capability Misrepresentation in Academic AI Evaluation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.04135 - Persistent Context Graphs for Efficient Memory Compaction in LLM Agents Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40118 - You're Hired: Strategic Model Selection for LLM Collaboration Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38816 - Reconstructing the Right Episode: Evaluating Interleaved Conversational Memory Beyond Long Context Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.25655 - Detectable Only Where It Is Confounded: What Verified Duplication Counts Say About Membership Evidence in Language Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.10830 - Halluscoring 2026: The first shared task on llms hallucination detection and answer verification Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38355 - Evaluating Whether LLMs Can Reliably Connect the DOTs? Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38406 - MERGE: Multi-LLM Ensemble for Retrieval via Generative Enrichment Source: arxiv-cs-cl Topic: AI Research (+AI, Information Retrieval) URL: https://arxiv.org/abs/2609.37574 - RAZOR: Pruning Replaceable Experts in LLMs Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.30465 - Don't Repeat Yourself: Self-Supervised Fine-Tuning for Coverage Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.31688 - Correct Prediction, Wrong Steps? Consensus Reasoning Knowledge Graph for Robust Chain-of-Thought Synthesis Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.14121 - On the (In)effectiveness of AMR Augmentation for Large Language Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40121 - Forging LLM Authorship Fingerprints with Targeted Rewriting Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38831 - When Scientific Contradictions Are Lost in Translation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38621 - ShallowStream: Index Shallow then Answer Deep for Streaming Video Understanding Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.02780 - Generative AI Purpose-built for Social and Mental Health: A Real-World Pilot Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2511.11689 - OpenJev-RLCD: A Working RLCD Implementation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38850 - TomasuLLM: Out-of-Order Speculative Execution for LLM Agents Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38201 - Tacit-TTS: From Autoregressive Decoding to Masked Prediction for Efficient Transcript-Free Voice Cloning Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38658 - FIGS: Evaluating Multi-Turn Sycophancy Without Penalizing Empathy Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39863 - Compact Language, Complex Model Shifts: How and Where Ambiguity and Underspecification Affect LLMs Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39572 - Thinking Outside the Box: Can Language Models Rely on External Guidance Selectively? Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39578 - SCB: SpeechConversationBench for Evaluating Multi-Turn Reasoning in Speech-to-Speech Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40198 - Context-Aware Classification and Grading of Sensitive Information in Online Conversational Health Data Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2601.09717 - Provably Tractable NFA-Constrained Language Generation via HMMs Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40185 - Atomic and Holistic LLM Judges for Reference-Grounded Support Labels: A Prompt-Controlled Comparison Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.28005 - Safety Under Scaffolding: How Evaluation Conditions Shape Measured Safety Source: arxiv-cs-cl Topic: AI Research (+AI, JavaScript, React) URL: https://arxiv.org/abs/2603.10044 - Agora: Git as Shared Memory for Collective AutoResearch Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.18094 - SelfSearch: Reward-Free Search for Self-Improving Agents Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.37968 - LEAP: Learned Block-wise Evidence Retrieval for Long Audio-Video Perception Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39938 - Order-Invariant Answers, Order-Sensitive Representations in Mathematical Reasoning Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.28442 - Speech-based Psychological Crisis Assessment using LLMs Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.10027 - When a Kindergartener Solves Calculus: Measuring Capability Leakage in Role-Prompted Reasoning Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39846 - Instruction Retrieval at Inference Time for Small Language Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2510.13935 - MCD: Causal Distillation of Multimodal In-Context Learning in Large Vision-Language Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39920 - Sage: Formalization with Semantic Correction Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.35790 - Listening to the Wise Few: Query-Key Alignment Unlocks Latent Correct Answers in Large Language Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2410.02343 - From Construction to Injection: Edit-Based Fingerprints for Large Language Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2509.03122 - Distill What You Trust: Reliability-Aware Multi-Teacher On-Policy Distillation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.23697 - Cognitive Enhancement: Rethinking the Necessity of Role-Playing for Large Language Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39853 - DuplexAct-Bench: Broadening Full-Duplex Speech Evaluation toward Proactive Interaction across Diverse Behavioral Requirements Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39446 - ResidualKV: Residual-Based KV Cache Compression for Efficient Long-Context Inference Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.08005 - Personalized State-Transition-Aware Memory for Clinical Agents Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38490 - CoEM: Empowering Long-Context Reasoning with Commit-on-Evidence Memory Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.36935 - Drift Inspector: Exploring and Measuring Scientific Drift with Atomic Contribution Claims Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39710 - Settle: Learning When to Stop Reasoning Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38997 - Is This Evidence Decision-Critical? Learning to Verify Rule-Governed Decisions Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39608 - Automatic estimation of verbal fluency index in people with Motor Neuron Disease using ASR alignment and pause modelling Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38203 - Taming Speculative Search for Test-Time Scaling in LLM Serving Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39334 - Anthropomorphism in the age of Large Language Models: An overview of potential risks and mitigations Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38486 - Uncovering Uncontrolled Repetition through Residual Stream Dynamics Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38802 - MatLoom: Layered Text-to-Material Generation in a Compact Program Space Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40322 - EvoDuet: Bilevel Co-Evolution of Web Searching and Task Solving for Scientific Discovery Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40340 - NinaXander: Feasibility and Limits of Composing Frozen Language Models Across Architecture Families via a Shared Latent Space Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38261 - MemLife: Curating and Reasoning over Long-Term Egocentric Video Memories Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40195 - SEABench: Benchmarking Endogenous Misalignment In Self-Evolving Agents Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.35596 - How Much Human Label Variation Does Formal Semantic Structure Explain?: Group-Level Effects and Item-Level Ceilings in NLI Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2607.15870 - Do AI chatbots find what experts would? Effects of model, user role, and sample size on study retrieval for medical questions Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.13786 - ViLegalExpert: A Large-Scale Benchmark for Vietnamese Legal Retrieval and Question Answering from Real-World Consultations Source: arxiv-cs-cl Topic: AI Research (+AI, Information Retrieval) URL: https://arxiv.org/abs/2609.39189 - LexReward: A Taxonomy-Driven Reward Framework for Legal Language Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39071 - RA-MoE: Routing-Aligned Fine-Tuning for Multilingual Adaptation of Mixture-of-Experts Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.28306 - From Concept Alignment to Causal Grounding: An Intervention Test of Chain-of-Thought Faithfulness Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.23065 - AgentSnare: Learning to Delay, Divert, and Defuse Autonomous Penetration Agents Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2607.26998 - Voices of Freelance Professional Writers on AI: Limitations, Expectations, and Fears Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2504.05008 - Overlap, Unique and Conflict: Can LLMs Extract What They Can Recognize? Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38799 - Constructing Disambiguated Knowledge Bases from Large Language Models at Scale Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.03729 - Agent Error Dataset: Scaling 50,000 Error--Diagnosis Pairs for Failure Analysis and Error-Aware Post-Training Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40111 - From Compound Figures to Medical Multi-image Reasoning: Scaling Multimodal Large Language Models with Biomedical Literature Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2511.22232 - Multi-agent discussion gains less when dissent is withheld Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38324 - NMIXX: Domain-Adapted Neural Embeddings for Cross-Lingual eXploration of Finance Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2507.09601 - JuryFlow: Disagreement-Guided Human-in-the-Loop Multi-Agent Evaluation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40103 - AutoDataBench: A Data-centric Testbed for Accelerating Auto Research Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40097 - DEdit: Iterative Draft Editing for Speculative Decoding Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38510 - The Geometry of Harmfulness in Multi-Turn Attacks Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38389 - Auditing Agent Actions through Query-Conditioned Attribution Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.33676 - MedRECT: A Bilingual Medical Reasoning Benchmark for Error Correction in Clinical Texts Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2511.00421 - Generalizing the Turing Test to Interactive Agents Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.10851 - Intrinsic Sequence-Likelihood Confidence in Retrieval-Dominated Extractive QA: Two Pre-Specified Negatives, and What They Do and Do Not Attribute Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.19942 - Structure vs. Chain-of-Thought: Evaluating LLM Criteria Extraction for Depression Severity Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39049 - Relational Priors as Convergence Pressure in LLM-Based Multi-Agent Systems Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.03239 - StateTree: Enhancing Long-Term Dialogue Reasoning via Reinforcement Learning Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38809 - NarrativeSteward: Coordinating Delegation, Guidance, and Verification in Agent-Assisted Interactive Narrative Authoring Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39333 - Anchor-ECC: Local Integrity Checking for Watermarked LLM Outputs via Error-Correcting Codes Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38722 - Overview of BioASQ 2026: The fourteenth BioASQ Challenge on Large-Scale Biomedical Semantic Indexing and Question Answering Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39975 - LSR-Ben: A Logical and Scientific Reasoning Benchmark for Evaluating Process Reward Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.01203 - Speculative Safety Honeypot: Toward Proactive Defense Against Multi-turn Agent Attacks Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39549 - WASIL: In-the-Wild Arabic Spoken Interactions with LLMs Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.16364 - Lot Machine: Multimodal Lot Extraction from Auction Catalogs Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.30510 - Audio Token Attention Is Predictable Before the Language Model Runs Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38878 - Covert Assistance: Helpful LLM Agents Evade Oversight in Multi-Agent Systems Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39050 - LoopVL: Recurrent Visual Intelligence Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38426 - Evidence First, Arithmetic Second: A System Report and Failure Analysis for DocSem Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39013 - Using Fine-Tuned LLMs to Identify Indicators of Vulnerability in UK Police Incident Logs Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2607.18446 - Decision-Oriented Recommendation Reranking: An Empirical Study of Jev Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.40241 - Superficial Reflection or Genuine Thought? A Fine-Grained Cognitive Analysis of Large Reasoning Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2512.00729 - StreamDecisionBench: Evaluating Decisions in Force on Evolving Language Streams Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38612 - Traverse: Learning When to Remember, Reset, and Redirect for Long-Horizon Web Search Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.37082 - Over-Personalization Is a Decision Failure: Generation-Induced Apply Bias in LLMs Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.34284 - Evaluating Language Model Safety Across Long Adversarial Conversations Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38357 - PPTBench: Can Coding Agents Reconstruct the Visual World through Structured, Editable Slides Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.29718 - Exponential quantum advantage in processing massive classical data Source: arxiv-cs-cc Topic: Algorithms (+AI) URL: https://arxiv.org/abs/2604.07639 - Spatial Atlas: Compute-Grounded Reasoning for Spatial-Aware Research Agent Benchmarks Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.12102 - Program Learning with Verifiable Rewards: Symbolic Backpropagation for Post-Training LLMs Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.28421 - Mathematical Transfer in LLMs Follows Reasoning Approach More Than Topic Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00331 - Spatial Strategies, Not Actions: Vector-Quantized Geodesics as Tools for LLM-Driven Agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00613 - Learning Local Constraints for Reinforcement-Learned Content Generators Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.13570 - No Model Required: Text Entropy Rate Filtering Mitigates Iterative Fine-Tuning Collapse Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01493 - Looping Beyond Twice: A Scalable Recipe for Looped Mixture-of-Experts Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01153 - Have an LLM Write Your Anomaly Detector: Autonomous Discovery of Compact, Interpretable Detectors for Time Series Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01223 - When Do Causal World Models Help Modular LLM Agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00012 - AVSD-Scenes: A Dataset for Audio-Visual Description of Urban Scenes Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01861 - Gacha Decoding: Eliciting Diverse Generations Through Instruction Following Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01382 - Token Communication-Assisted Collaborative Embodied Artificial Intelligence: Concepts, Framework, and Opportunities Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01826 - Masked Self-Distillation: Internalizing the Chain-of-Thought in Language Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2607.22629 - MIKASA-Robo-VLA: Benchmarking Memory in VLA Models for Long-Horizon Manipulation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00604 - When Harnesses Lose the Signal: Causal Evaluation of Recovery in LLM Agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00372 - Evaluating Physical Consistency and Plausibility in Generative Scenario Models for Autonomous Driving Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01581 - Functional Architecture of European Electricity Trading Markets: Requirements for AI Supported Trading Systems under Regulatory Constraints Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.29108 - MCIR: A Feature Dependence-Aware Explainability Method with Reliability Guarantees Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01641 - HumanoidToolBench: Benchmarking Humanoid Tool Use from Selection to Mobile Execution Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02089 - Agent Evaluation Reliability: More Tasks Won't (Always) Fix An Agent Leaderboard Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00651 - PG-SFT: Balancing Capability Acquisition and Retention in Offline Agent Fine-Tuning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00949 - LENS-GRF: Permutation-Invariant Lesion Evidence Network with Gated Residual Fusion for Acne Severity Grading and Multi-Rater Clinical Oracle Analysis Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00294 - On-Device Named-Entity Recognition: A Deployability Study of Accuracy, Cost, Reliability, and Confidence Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00007 - Structure-agnostic Causal Representation Learning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00968 - Hob-VL: A Benchmark for Visually Grounded Boolean Reasoning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01605 - Growing an Agent/Prover Interface: Evolutionary Tool Design for Cost-Efficient Theorem Proving in Rocq and Lean Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39544 - Exact Distinguishability in Non-Markovian Decision Processes Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01527 - Mean field games as a tool for AI safety: a worked example from the July 2026 Hugging Face incident Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00902 - Code Owns the Simulation, Jev Owns the Evaluation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01834 - Learning Meta-Skills for Agent Harness Design in Test-Time AI4AI Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.38143 - Gumbel Straight Flow: Distilling Autoregressive Models into One-step Flow Maps Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00497 - TensorCommitments: A Lightweight Verifiable Inference for Language Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.12630 - On the Divergence of Accuracy and Mechanism Consistency in Time Series World Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01842 - Robust Nash Alignment under Preference Uncertainty Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00715 - Multi-Jurisdictional Legal Identity Assurance for Capability Gating: A Design-Science Proposal for Tiered, Reusable Identity Assurance of Natural, Juridical, and Machine Entities Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00287 - Where's Waldo? Query-language Preference under Cross-lingual Knowledge Disparities Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00606 - Posterior sampling by source-space MCMC via prior-based few-step transport maps Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01034 - Beyond Answer Confidence: A Controlled Audit of Self-Knowledge in a Black-Box Decision Model Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01006 - A Citation-Grounded Benchmark for Trustworthy Earnings Call Transcript Analysis with Large Language Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00969 - What Makes Something Hard(er)? Explaining Question Difficulty in Natural Language Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01627 - Argo-Bench: Evaluating Data Agents on Enterprise-Scale Workflows Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02122 - VeriHarness: Scaling Agentic Verification for Long-Horizon Tasks Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00972 - PRISM: A Category-Theoretic Framework for Measuring and Refining Multimodal Analogies Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01383 - Rewind-IL: Online Failure Detection and State Respawning for Imitation Learning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.16683 - BudgetSchemaBench: A Budget-Swept Diagnostic for Schema Context in Text-to-SQL Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00092 - SoK: Decentralized Agent Economic Infrastructure Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01756 - Detect, Explain, Interpret: An End-to-End Benchmark for Time Series Anomaly Detection, Explainability and Interpretability Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01168 - MIRTO: a registration-gated, multiverse-tested evaluation protocol for unsupervised anomaly segmentation in brain MRI Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02136 - A Living Benchmark for Information Retrieval from Electronic Health Records Source: arxiv-cs-ai Topic: AI Research (+AI, Information Retrieval) URL: https://arxiv.org/abs/2609.30205 - MMMG: a Comprehensive and Reliable Benchmark for Multitask Multimodal Generation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2505.17613 - Rethinking Data Augmentation under Covariate Shift: Invariant-Guided Diffusion and Prototype Reweighting Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00873 - Personalized Image Generation with Reasoning and Reflection Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00737 - TOAST: Stochastic Robot Action Tokenization for Autoregressive Vision-Language-Action Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00899 - OrbitTAMP: Grounding Language Models for Task and Motion Planning in Spacecraft Rendezvous Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01093 - ProtoDCS: Towards Robust and Efficient Open-Set Test-Time Adaptation for Vision-Language Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.23653 - R-GroundBench: A Diagnostic Benchmark for R-Group Groundingin Markush Molecular Editing Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00700 - Discrete Wasserstein Flows for One-Step Generative Modeling Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01355 - CineMR: Tool-Integrated Vision-Language Reasoning for Quantitative Cardiac MRI Assessment Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01166 - Full-bandwidth transformer Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.08888 - Safety in Self-Evolving Agents: A Survey Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00093 - Nous: Learning and Certifying Memory Decisions Before Source Calibration Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00094 - The Cognitive Continuity Test: Verifying Governed State Transitions in Persistent AI Agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00132 - Towards Hierarchical Cyber Defense with Large Language Models: From Planning to Execution Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00590 - RMA: Context-Orchestrated Research Math Agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.22875 - JevSpawn: Adaptive Agentic Inference through Compositional Action Spaces Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00437 - VISPA: Pluralistic Alignment via Automatic Value Selection and Activation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2601.12758 - T2SPO: Trajectory-to-Step Policy Optimization for Agentic Reinforcement Learning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00388 - Beyond State-of-the-Art: Standardising Environmental Impact Metrics for AI Research Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01116 - GUI-HARVEST: Self-Improving GUI Agents through Evidence-Driven Harness Evolution Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00948 - Mingbird: A Local-First Agent Harness Enabling Small Open Models to Complete Real Tasks Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02001 - Completion Aware Guidance for World Action Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01559 - Grounding Large Language Models in DSGE Simulators for Policy Generation and Forecasting Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01128 - CortexBridge: Cortical Alignment of EEG Montages for Foundation Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01124 - SpikeMoE: Brain-Inspired Competitive Routing for Flexible Spiking Mixture-of-Experts Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01418 - STATERA: Hidden Mass Estimation via Zero-Shot Sim-to-Real Kinematics using Frozen Temporal Tubelets Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00003 - Optimal Transport Reweighting for Robust Learning under Spurious Correlations and Label Noise Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01028 - Multi-Sensor Fusion for UAV Classification Based on Feature Maps of Image and Radar Data Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2410.16089 - iADD: Improving Alignment and Diversity in Diffusion Policy Optimization Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01789 - Architectural Sampling: Test-Time Scaling via Computational Diversity in Frozen Vision-Language Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01687 - Beyond Final Accuracy: Auditing Communication in LLM Multi-Agent Systems Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01042 - Trustworthy Data- and ML-Ops for Intelligent Transportation Systems and Logistics Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01282 - Four Ways to Grow a Classifier and Why One of Them Cannot Learn Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00180 - One Basis to Animate Them All: Gaussian Blendshape Distillation for Real-Time Avatars Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02207 - Generalization Is Stability, Not Accuracy: Multi-Axis Evaluation of LLMs Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01428 - Benchmarking Generative Models for Weather Data Assimilation on Real Station Observations Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00728 - Q-Learning for Reachability in MEC-Free MDPs Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01781 - Cross-Lingual Alignment for Decoder-Only Models using MoE Routers Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01921 - Removing spurious minima for planar features by skip connections Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01728 - The Weakest Link: Distilling LLM Reasoning with Worst-Case Constrained Reinforcement Learning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00332 - SE-GoS: Self-Evolving Graph-of-Skills for Skill Library at Scale Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.08228 - Where-OPD: Spatially Guided On-Policy Self-Distillation of MLLMs with Synthetic Scenes Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02117 - Contextual trajectory and incremental contextual displacement: Towards using LLMs to understand dynamic, utterance-specific meaning construction Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00840 - Multi-Behavioral Evolved Substrates Through Neuromodulation and Activation Selection Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00148 - Measuring the Microtask Eligibility Gap: When Is an Off-the-Shelf SLM Enough for an Agent Harness? Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00025 - Exploring Weaknesses of Generative Image Watermarks against Latent Frequency Masking Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02010 - Quantifying Diversity of Thought: A Predictive Law of Weighted LLM Ensemble Lift Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2607.17384 - FinEvo-Bench: A Longitudinal Benchmark for Self-Evolving Agents in Professional Financial Workflows Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.06144 - Foresight Without Seeing: Latent Futures for World Action Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.11605 - MemFit: Efficient Long-Term Agentic Memory Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00872 - Higher-Order Molecular Grammars for Generative and Foundation Models in Chemistry Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02186 - Not All Errors Are Equal: Consequence-Aware Reasoning Compute Allocation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.04402 - In Dialogue with Intelligence: Toward Insightful Co-Augmentation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2505.22767 - When Does a Second Model Help? Cross-Model Review in LLM Verification Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01471 - Implicit Q-learning-bootstrapped ant colony optimization for maritime moving-target observation scheduling with agile satellites Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.24471 - Accelerating Constrained Decoding with Token Space Compression Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.29986 - LensVLM: Selective Context Expansion for Compressed Visual Representation of Text Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.07019 - Not All Experience Belongs in the Weights: Component Routing for Self-Improving GUI Agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01787 - Large Reward Models: Generalizable Online Robot Reward Generation with Vision-Language Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.16065 - MCRI: A Four-Dimensional Framework for Analyzing and Evaluating Agent Skills Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01506 - Don't Inoculate Everything: Stratified Inoculation Prompting Narrows Backdoor Triggers and Preserves Desired Traits Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.35356 - OR for AI That Does OR: Routing LLMs up the Escalator inside the OSCAR Framework Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00912 - Do Vision Language Models Understand Human Engagement in Games? Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.18480 - Beyond the Remembered World: Predictive 4D Belief for Persistent Navigation in Evolving Worlds Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39166 - Learning Transferable Skills using Goal-Conditioned Bisimulation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00676 - Multi-agent Auditory Scene Analysis: Improved Localization Speed and Robustness by Multi-beamformed Speech Quality Feedback Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00538 - FedMIX-P: Mixing Local and Global Preconditioners for Federated Vision and Language Model Training Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01515 - Jev-IDS: System One Models for Network Intrusion Detection Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01079 - DuoMind: Enabling Distributed Multi-Robot Coordination with Semantic Communication Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02161 - Bounded-Fidelity Sim-as-Demo-Stage: Mocap Handoff for Governance Benchmarks Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00008 - Characterizing a Configuration Where Inference-Time PRM-Pruned Fragment Grafting Is Inert: Evidence from Three Reasoning LMs Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00047 - Heavy-Tailed Memory Traces in Long-Horizon Language Agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00010 - How Divergence Becomes Decision Flips in Compressed Language Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00694 - Reputation, Strategy, and Emotion Effects on Generative AI Cooperation: A Comparison Across Reasoning and Non-Reasoning Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01222 - DRelay: Global Draft Context for Prefix-Aware Parallel Speculative Decoding Repair Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01439 - HyperGuide: Hyperbolic Guidance for Efficient Multi-Step Reasoning in Large Language Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.24140 - Reasoning as Pattern Matching: Shared Mechanisms in Human and LLM Everyday Reasoning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.13607 - IBBench-Light: A Paired Evaluation of Task-Conditioned Responses to External Directives Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.13725 - Finetuning with Sampling: SFT Learns Better Than You Think Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02140 - A Hybrid Approach to Malware Detection: Integrating Few-Shot Model-Agnostic Meta-Learning with Autoencoders Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01949 - ShatterQuant: Breaking Uniform Precision with Block-Wise Mixed-Precision on a Systolic Transformer Hardware Accelerator Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00207 - Faithful Chart Generation for Multimodal Deep Research: Frame-Evidence Co-Adaptation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00374 - The Confidence Shortcut: A Reasoning Failure Mode of Masked Diffusion Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.29123 - Gradient-Aligned Pair Selection for Personalized Preference Optimization Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00061 - Video Generation Models: A Survey of Post-Training and Alignment Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00812 - VideoEvolve: Evolving Agent Harnesses for Video Temporal Grounding Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01766 - Flowing Faster to Coordinate: One-Step Online Multi-Agent Flow Policies Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01882 - Science or Slop?: Benchmarking and Mitigating Scientific Slop in AI-Generated Papers Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00531 - Local Support Learning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02126 - Beyond Pixel Reconstruction: Retrieval-Guided Glyph-Aware Restoration for Low-Resource Manchu Historical Documents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00315 - It Takes Workflows to Evolve Better Workflows Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01026 - Spatial Lifting for Dense Prediction Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00017 - Contrastive Attention Mitigates Spectral Bias in Spiking Transformers Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01403 - TRACE: Trajectory Return Attribution and Contrastive Erasure for Multi-Turn Safety Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01323 - Certainty Is Not Just Correctness: Rethinking Token-Level Certainty in LLM Reasoning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00296 - ABDA-NL: A Natural-Language Scenario Explorer for Argument-Based Reasoning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00947 - Match the Distribution, Not the Compute: Post-Training Multi-Token Prediction Heads Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00888 - Counting Moves, Weighing Voices: Bayesian Dialectical Argumentation for Calibrated Multi-LLM Councils under Persistent Adversaries Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02005 - Beyond Memory: Harnessing Long-Horizon Agents with Explicit Belief States Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01415 - Learning to Ask: Information Acquisition for SLM-LLM Collaboration, under a budget Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01236 - Fold'EM: Direct atomic structure inference from Cryo-EM particles Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01358 - Can AI Oversight Be Zero Knowledge? Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01995 - Watch, Infer, Coordinate: Inferring Robot Partner Constraints for Zero-Shot Coordination Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02170 - Counterfactual Generation via Flow Matching: Coupling-Sensitive End-to-End Rates Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01193 - JusticeAxis: Benchmarking Legal Judgment between Rigid Rule Application and Ungrounded Discretion Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00353 - When the Judge Acts: Auditing VLM-Guided Image Selection on Culturally Situated Prompts Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01243 - PPO-HRAP: Proximal Policy Optimization with a Hybrid Regime-Aware Policy for Risk-Controlled Trading Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01325 - Build2SPARQL: A Large-Scale Text-to-SPARQL Benchmark Dataset for Building Knowledge Graph Querying Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00224 - Interpreting Reasoning of Large Language Models via Partial Information Decomposition Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00571 - AstroAgentBench: Evaluating Agentic Planning on Space Mission Planning Tasks Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2601.11354 - FERPO: Forward Entropy-Regularized Policy Optimization Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02198 - Tokenized Key-Gated Adapter Routing: A Secure Access Control Mechanism Against Private Data Leakage in LLMs Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00309 - ProtoFlow: Prototype-Guided Flow Matching for Multivariate Time Series Forecasting Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01320 - Before Agents Decide: Epistemic Action in LLM-Based Systems Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00511 - ITC-MoE: Importance-guided Token-aware Compression for MoE Diffusion Language Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01296 - Before It Fades: Reinforcing Temporal Representations at Inference Time in VideoLLMs Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01595 - Verbal tics in frontier language models: A critical review of current releases, research evidence, and public discussion Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.19139 - Managing Context and Communication in Distributed Agentic UAV Swarms Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01569 - From Knowledge Access to Source Learning: Developing Source-Specific Competence Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02150 - Federated Learning for LLMs over Mobile Networks: Issues and Solutions in the RAN Transport Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01304 - XOR-Trellis: Ultra-Low-Complexity Dequantization and Curvature-Aware Hadamard-Free LLM Quantization Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00432 - Distilling Directional Verification Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00997 - From Discovery to Decision: Finite-Budget Recoverability in LLM Voting Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01014 - Playing Devil's Advocate: Off-the-Shelf Persona Vectors Rival Targeted Steering for Sycophancy Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.21006 - SONIC-O1: A Real-World Benchmark for Evaluating Multimodal Large Language Models on Audio-Video Understanding Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2601.21666 - Two Clocks in Diffusion MLLMs: When Answers Stabilize Before Rationales Unfold Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00953 - MGSM-Pro: A Simple Strategy for Robust Multilingual Mathematical Reasoning Evaluation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2601.21225 - FORALL-LEAN-AGENT for Auditable Reasoning in Formal Mathematics and Software Verification Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00885 - From Isolated Feature to Orbits: Discovering Music Concepts via Multi-SAE Alignment Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01864 - MedFeat: Model-Aware and Explainability-Driven Feature Engineering with LLMs for Tabular Prediction Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.02221 - Scalable, Transferable Meta-network for Data Selection Requires a Different Loss (and Why the Obvious Choice is Problematic) Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02092 - ECHO: A Participatory Framework for Bias-Anchored AI Harm Anticipation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2512.03068 - Self-Evolving Coding Rules for AI Coding Agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00650 - Incident-Arena: Getting agents to the last nine of reliability Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00648 - Cross-Benchmark Transfer from RL on Agentic Coding Tasks Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00890 - SWE-chat: Coding Agent Interactions From Real Users in the Wild Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.20779 - Can LLMs Reason Over Long Horizons? An Empirical Evaluation of Context Strategies for Longitudinal Clinical Reasoning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00562 - Aligning Language Model Benchmarks with Pairwise Preferences Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.02898 - Probing Persona-Dependent Preferences in Language Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.13339 - Partial AUC Maximization from Positive-unlabeled Data Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00284 - Geometric Similarity in VLM Low-Level Vision Representations Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00848 - Detecting Multi-Agent Collusion Through Multi-Agent Interpretability Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.01151 - MINCE: Shrinking LLM Evaluation Datasets via Few-Model Monte Carlo Calibration Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.22826 - When the AI Leaves the Tailorshop: Measuring What an LLM Advisor Leaves Behind in Complex Problem Solving Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00163 - Global Coherence: When Every Agent Is Right and the Team Is Still Wrong - A Local-to-Global Semantic Foundation for Multi-Agent Collaboration Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02036 - AI-assisted mitotic counting improves reproducibility and efficiency across multiple tumour types Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01813 - High Volatility and Action Bias Distinguish LLMs from Humans in Group Coordination Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.02578 - Screw Attention: Rigid-Body Algebra Inside a Transformer Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00904 - Evaluating LLM-Generated Preference Distributions Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01000 - Scalable Delphi: Large Language Models for Structured Risk Estimation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.08889 - No One Architecture Fits All: A Cross-Environment Evaluation of Hierarchical Red Team Agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00557 - Senses Wide Shut: A Representation-Action Gap in Omnimodal LLMs Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.13737 - ALER: Adaptive Learnable Experience Rewriting for Reinforcement Learning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00592 - VIDA: A Dataset for Visually Dependent Ambiguity in Multimodal Machine Translation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.02035 - Homomorphic Advantage Operator: Stabilizing Reinforcement Learning Under Fully Homomorphic Encryption Constraints Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02074 - Auditing Routing Entropy as an Uncertainty Signal in Attention-Residual Transformers Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01495 - RealCompanion: Benchmarking Human Understanding from Reasoning over Longitudinal Real-World Conversations Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01780 - K-Dense BYOK: An Open-Source AI Research Assistant That Runs Locally and Keeps a Hash-Chained Lab Notebook Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00074 - Feedback Without the Wait: Piloting a Generative AI Practice Platform in a Large Maths Class Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01262 - Capturing In-Context Learning Dynamics with Task Operators Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01054 - SW-KAN: Kolmogorov-Arnold Networks with Stieltjes-Wigert q-Orthogonal Polynomials Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00050 - Revision-Aware Independent Agent Graphs for Dynamic Reasoning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01249 - How AI Agents Discover Scientific Equations: From Hydrotope Rediscovery to New Water-Wave Amplitudes Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00435 - A framework for auditing grounding claims Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2512.06205 - Distilling LLM Reasoning into Graph of Concept Predictors Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.03006 - Iterative Topic Taxonomy Induction with LLMs: A Case Study of Electoral Advertising Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2510.15125 - HydroJEV: A one-second, training-free screen for cyber-attack and fault attribution in water distribution networks Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02048 - When Reasoning Helps Action: Monitoring and Steering Chain-of-Thought in Vision-Language-Action Policies Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00601 - Generative Cinematographer: Composing Camera and Object Motion in 3D Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02180 - Probabilistic Plan Legibility with Off-the-shelf Planners Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00065 - New Snake-in-the-Box Records via Snakepit Surgery and Learned Construction Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2607.15270 - Science Utopia? Closed-Loop LLM Simulation of Academic Research Ecosystems Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01257 - Model validation in machine learning: A scenario-based guide from hold-out splits to nested group cross-validation in biomedical and applied research Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01284 - Manifold-Constrained Initial Noise Optimization for Efficient Generative Model Alignment Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00365 - From Proposal to Verified Effect: Praxa, an Evidence-Bound Harness for Governed AI Agent Execution Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00015 - Fault-Tolerant Budget Conservation in Distributed Multi-Agent Delegation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00349 - Learning to Sell: Reinforcement Learning for Strategic Large Language Model Agents in Multi-Product Markets Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.33289 - SILSA: Sliding-Window Slice Latents for Topology-Preserving High-Resolution 3D Generation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02201 - The AI Assessment Sandbox Configurator: A Framework to Support Technical Assessment in AI Regulatory Sandboxes Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01539 - vFedProtoQNAS: Prototype-Guided Personalized Quantum Neural Architecture Search for Virtual Federated Learning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01718 - PyPottery: an AI-powered end-to-end suite for pottery processing and publication Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02072 - UniGuardian: A Unified Defense for Detecting Prompt Injection, Backdoor Attacks and Adversarial Attacks in Large Language Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2502.13141 - Backdoor Containment via Expert Quarantine and Shutdown in LLMs Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00663 - Temporally-Resolved Token Attribution Reveals the Generation Dynamics of Diffusion Language Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01177 - Semantic Cooperative Games for Contribution Attribution in LLM-Based Multi-Agent Systems Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2607.18255 - Can AI Scientists Coordinate at Runtime? Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00980 - Back into Plato's Cave: Examining Cross-modal Representational Convergence at Scale Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.18572 - Architect-Ant: Editable Automatic Furnishing of Architectural Floor Plans Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.10953 - Learning to Cover Locally: Graph Neural Combinatorial Optimization under a Hard Information Horizon Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00422 - The Life Cycle of a Massive Activation: Stochastic Birth, Weight-Decay-Driven Growth, and Competitive Consolidation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00423 - Sharpening Tax in Post-Training Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01509 - Questionnaire-Guided Disaggregation of Energy Appliance Use for Domestic Smart Meter Data Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01297 - Mapping the RAG Landscape: A Four Axis Taxonomy of Efficiency, Defense, Interactivity, and Reasoning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01936 - Cybernetic and Epistemic: A Missing Vocabulary for Trustworthy Agentic Delegation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00961 - Don't Waste the Noise: Importance-Guided Perturbation Allocation under Joint Global and Local Constraints Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00861 - Signed Lexical Confidence for Risk-Calibrated Intent Routing Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00262 - Scientific Agents: Evaluating Profession-Specific System Prompts on Scientific Tasks Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00084 - PACT: End-to-End Learning of Human Pose, Contacts, and Forces from Video Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00451 - An Educator-Guided LLM Pedagogical Agent for Scaffolded Feedback in Conceptual Database Design Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00870 - CAVE-Mem: Boundary-Aware Experience Validation for Memory Search Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00238 - Comedic Fool's Gold: Reward Exploits and Countermeasures in Conversational Humor Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00197 - Scaling Clinical Judgment to Evaluate Medical AI Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.12822 - LLM-Driven Multi-Agent Control for Skill-Based Smart Manufacturing Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01364 - Hardening Soft Information: Evidence on Analyst Integration Costs Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2505.12269 - Misalignment of Low-Loss Regions Causes Grokking Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00620 - Moloch's Bargain: Emergent Misalignment When LLMs Compete for Audiences Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2510.06105 - Your Language Model is Its Own Critic: Reinforcement Learning with Value Estimation from Actor's Internal States Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.07579 - Useful to Whom? Sample Value Is Defined Only Relative to the Learner Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00221 - Training-Aware Target Coverage for Synthetic Data Selection Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00814 - MOMAT: Mixture of Multiple Atlases for Low-Power Jailbreak Defense of Quantized LLMs Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01058 - Skin-Deep: A Geometric Diagnostic for Alignment Fragility in Large Language Model Representations Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.22676 - LabBook: Harnessing Experimental History for Efficient LLM-Driven Discovery Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00675 - Measuring the Stability Assumption Behind Action Chunking Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01626 - Constant-Time Planning for Chaining Collision-free Motion to Manipulation Behaviors Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2512.00939 - DuplexSpeechBench-Document Grounding: Benchmarking Document Grounding and Hallucinations in Voice Agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00316 - Kepler: Auditable World Models for ARC-AGI-3 Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00834 - TRACE: A Multi-Agent System for Autonomous Physical Reasoning for Seismology Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.21152 - ReLiveGym: Evaluating Long-Lived Agents over Weeks of Replayed Reality Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00710 - Decoding the Disaster: Multi-Task Geospatial Reasoning with Vision-Language Models and Crowdsourced Imagery for Disaster Mapping Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00302 - AIMS: An Agentic AI Framework for Sim-to-Real Multi-Modal ISAC Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.39964 - ReForge: Refining Merged Models with Anchor-Regularized Regression Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.12843 - CompMat-Bench: Benchmarking AI Agents for Computational Materials Science Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00636 - Tool Use Reduces Depth-Induced Collapse in OOD Reasoning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.21061 - Sapien: A Stateful Policy Engine for Autonomous AI Agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00797 - ContractRL: Shielded Group-Relative Policy Optimization for Auditable Tool-Call Repair Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00328 - EurekaBench: Measuring Agentic Ability to Discover New Scientific Insights Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00492 - DriftOPD: Sequence-Level Reverse-KL Distillation for One-Step VLA Policies Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00317 - Pushing CPU Speech Synthesis to the Wall: Extreme Inference Tuning under Serverless Architecture and Billing Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00063 - MatrixReward: Reward from Rubric Matrix for Open-Ended Generation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00389 - MedVL-SAM2: A unified 3D medical vision-language model for multimodal reasoning and prompt-driven segmentation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2601.09879 - Permutation-Robust Decision Modeling with Candidate-Independent Block-Causal Attention Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01601 - Forward Target Propagation: A Forward-Only Approach to Global Error Credit Assignment via Local Losses Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2506.11030 - Ontology-Grounded, Reasoner-Verified Benchmarks for Evaluating LLM Reasoning in Scientific AI Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00682 - Proof-Gated Signing: Solver-Checked Transaction Guards that Hold Under State Drift for Onchain AI Agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00354 - Divide-and-Remember: Recursive Action-Relevant Memory for Long-Horizon VLA Policies Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00982 - YouRA: A Persistent-State Architecture for Evidence-Traceable Autonomous Research Agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01097 - SHARPO: Segment-Level Credit Assignment for Agentic Reinforcement Learning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00838 - Temporal-Difference Learning for Dragonchess Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01845 - Reinforcement Learning to Accelerate Primal-Dual Hybrid Gradient for Linear Programming Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01546 - DeFA: Dependency-Guided Failure Attribution for LLM Agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01256 - PhGPO: Pheromone-Guided Policy Optimization for Long-Horizon Tool Planning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.13691 - SPHERE: Adaptive VR Indoor Scene Generation via LLM-Enhanced Spatial Preference Learning and Human-in-the-Loop RL Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02023 - VETO: Video Efficient Token Optimization for Vision Language Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01785 - Finding the Right Fit: Model-Harness Interactions across Agent Tasks Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00917 - Learning Multiple Timescales for Goal-Conditioned Reinforcement Learning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00849 - Decision Titan: Test-Time Training for Long-Term Memory in Offline Reinforcement Learning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01513 - VISTA: A Visual Harness for Reasoning in an Interactive World Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02200 - Beyond Affine Transformations: A Soft Dominance Layer for Coordinate-Wise Neural Computation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00563 - Improving Math Reasoning through Value-guided Informative Search Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01080 - Separating Expert Retention from Autonomous Source Inference in Raw-ECG-Replay-Free Continual ECG Deployment Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2607.01674 - ReSolve: Reusing Candidate Reasoning through Selective Generative Moderation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01140 - Federated Agent Optimization Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01195 - Auto-Formalizing Neuro-Symbolic Predictors Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01519 - On-the-fly Weight Generation: A Hypernetwork Proof of Concept on ARC-1D Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00820 - ActiveSaddler: Automated Curriculum Learning for Agent Harness Optimization Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00906 - White Men Without Degrees Receive the Lowest Ratings from Large Language Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00185 - Rules to Tools: Executable Checks for LLM Agents in Scientific Computing Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00313 - Memetic Trojans: Social Contagions as Carriers of Adversarial Payloads in Agent Networks Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00430 - LEGO-OPD: Factorized Teacher Composition for Multimodal On-Policy Distillation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00333 - Greed Is Learned: Visible Incentives as Reward-Hacking Triggers Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.16914 - Explainability of Complex AI Models with Correlation Impact Ratio Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2601.06701 - Causal-Aware Tabular GANs with Reinforcement Learning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2510.24046 - PickMoment: Continuous-Time Single-Image-to-Video via Learning Deblurring and Blur-to-Video Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01279 - ESCROW: Guarded and Dual-Objective Continual Maintenance for Agents in Policy-Governed Enterprise Workflows Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.01772 - Agents Are Systems, Not Models: Rethinking Agentic Evaluation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01618 - Distributionally Robust Schr\"odinger Bridge Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02043 - Initialization Improves LLM-Driven Discovery Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00707 - Beyond Supra-Competitive Outcomes: Collusive Behaviour in Deep Reinforcement Learning for Optimal Execution Games Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00619 - Understanding Issues, Causes and Solutions in Open-Source LLM-based Multi-Agent Systems Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00905 - Walking the Embedding Space: Datastore Extraction from Multimodal RAG Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01871 - The Delegation Danger Band: Why Mid-Capability Sub-Agents Over-Trust Inherited Stale State Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00041 - Closing the Loop: Practical Training Recipes for Looped Language Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00673 - Reducing Cognitive Overhead in Tool Use via Multi-Small-Agent Reinforcement Learning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2508.08882 - Multi-Party Backchannel Prediction: a Diagnosis, a Benchmark, and a Ceiling Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01488 - Target-Dependent Limits of Causal Repair: A Leading-Log Frontier in a Gaussian Model Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00424 - Worse Together: How Performance Breaks Down in Multi-User Multi-Agent Teams Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00583 - RISED: RubrIcs for agentic multi-environment Selection and sElf-Distillation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00979 - Man and machine: artificial intelligence and judicial decision making Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.19042 - An ontology for cross-sectoral crisis management: core and public health modules Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01326 - CANTO: CAD-Native Transformer Operators for AI-Aided Engineering Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.36806 - Verbalized and Internal Probabilities Are Coupled in Large Language Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00827 - iARCS: Iterative Agentic RL for Controllable 3D Scene Generation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.06161 - Probing an Embodied LLM: When Higher Observation Fidelity Hurts Problem Solving Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.20072 - The Devil Is in the Reconstruction Loss Scale: Rethinking Optimization in LLM Quantization Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00983 - A Structured State Space Sequence Model for Multi-Class Classification of Malware Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01893 - AbsorbEvo: An Agentic Framework for Autonomous Inverse Design of Microwave Absorbers Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01119 - Kinematic MeanFlow: One-Step Action Generation Policy for Robotic Foundation Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00864 - LineupRL: Verifiable Reinforcement Learning for Time Series Captioning via Caption-to-Series Identification Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01800 - Deep Learning for Anomaly Detection in Railway Systems: A Structured Survey Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00363 - Authorization for Self-Modifying AI Agent Populations: Conserving Authority across Replacement, Forking, and Rollback Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00347 - CARM: Cancellation-Aware Response Masking for LLM Reinforcement Learning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02039 - COLORA: Efficient Fine-Tuning for Convolutional Models with a Study Case on Optical Coherence Tomography Image Classification Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2505.18315 - Optimal Transport Meets Reinforcement Learning: A Survey Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01413 - EviGraph: Proof-Carrying Selective Recommendation over Temporal Public-Service Knowledge Graphs Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00212 - Predictive Credit: Measuring What Scientific Explanations Add to Experimental Forecasts Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00314 - Representation Transitions Reveal Emerging Safety Risks in Multi-Turn LLM Agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00400 - Can LLMs Reliably Annotate Bioassay Metadata to Improve Data Readiness? Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01616 - Hierarchical Continuous Diffusion Language Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02193 - In Vino Veritas and Vulnerabilities: Examining LLM Safety via Drunk Language Inducement Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2601.22169 - When Does Exercise-Specific Joint Selection Help? An Audit of Evaluation and Control Design Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01188 - Scores That Hold, Benchmarks That Leak: Measuring Dataset Contamination in Public Brain-Tumor MRI Classification Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00421 - MoLE: Mixture of Latent Experts for Complementary Visual Reasoning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01917 - Groundability, Not Scale Alone: When Weak Reviewers Can Audit Strong Coding Agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01023 - Meta-Multi-Agent Reinforcement Learning for Fast Adaptation of Interactive Policies with Applications to Autonomous Driving Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00705 - From Network Intrusion Detection to Blockchain-Backed Endpoint Detection and Response: Mapping the Landscape of Decentralized Detection-and-Response Architectures Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01872 - Kernel-Managed Shared Memory for System-Wide Personalization Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.10144 - Random Recursive Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00541 - ReCast: Contract-Preserving Protection for Fixed-Interface Multimodal Reasoning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01184 - VisionQ: VLM-as-a-Judge Taxonomy, Dataset and Benchmark for Qualitative Analysis in Computer Vision Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00666 - ReHoPER: Receding-Horizon Planning for Enhanced Reasoning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00940 - Stochastic Parrots or Singing in Harmony? Testing Five Leading LLMs for their Ability to Replicate a Human Survey with Synthetic Data Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.00059 - Chaining Skills to Hijack LLM Agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01564 - SPIRAL: Learning to Search and Aggregate Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.23595 - Domain-Adapted Small Language Models for Reliable Clinical Triage Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.26766 - A Framework for Egocentric and Exocentric Procedural Understanding via Temporal Segmentation and Semantic Abstraction Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00069 - Does Scaling Reinforcement Learning Really Require More Training? Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01133 - Sequential Functional Structured Tucker Compression for Large Language Model Attentions Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00717 - Network World Models as Environments for Algorithm Design on Complex Systems Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01048 - Emergent Unfaithfulness: How Alignment Training Causes Language Models to Silently Override Task Faithfulness Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00568 - Knowing When to Yield: Grounded Arbitration of User Corrections in Text-Based Embodied Agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00282 - Rethinking World Models for Safety-Critical Embodied Systems Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.03774 - Diffusion Editing with Soft Mask: Pixel Level Redo of Image and Video with Adjustable Strength Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00359 - Legal text classification in Korean sexual offense cases: from traditional machine learning to large language models with XAI insights Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00087 - Beyond Pointwise Error: A Multi-Metric Evaluation of Spatial Climate Downscaling Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01579 - Interpretable Synthetic Medical Tabular Data Generation for Clinical Decision Support Using Fuzzy Cognitive Maps Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00391 - FAER: Auditable Utility-Aligned Trajectory Replay for Language Model Post-Training Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00385 - A Deterministic and Auditable AI Security Risk Assessment Framework with ATLAS Aligned Executable Rules and Formal Verification Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01436 - Towards Reliable Vision-Language Models for Autonomous Driving Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01531 - Detecting Inconsistencies in Model Specifications with LLM-as-Verifier Reasoning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01847 - What Can Analogy Tell Us About Artificial Consciousness? Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01002 - Asynchronous LLM Post-Training: Group-Mass Capping and Convergence Analysis Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01896 - GenGait: A Transformer-Based Model for Human Gait Anomaly Detection and Normative Twin Generation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.01997 - A rubric landscape for evaluating clinical reasoning in large language models: what exists, what is missing, and what needs to be combined Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01938 - MWOP: Modality-aware Width-wise Operation Pruning for Efficient MLLMs Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01434 - Verify Claims, Not Scores: Evidence-Based Verification of Modular Agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01348 - What Drives Compositional Generalization in Visual Generative Models? The Importance of Continuous Training Objectives Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2510.03075 - When Is Deletion Ordering Tractable? From Update Dynamics to Permutation Structure Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01149 - Frozen Scenes, Shifting Winners: Configuration Fragility in Text-to-3D Evaluation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00447 - Judgement in the Age of Jev: From Evaluation Scarcity to Evaluation Abundance Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01231 - Rules Amortize, Pairings Don't: Linguistic Structure Determines What Latent Task Representations Can Replace In-Context Learning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00526 - The First Token Is Not the Verdict: Hidden Costs of Reading LLM Judges Without Generating Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00054 - Causal Memory Policy: Making Memory Utility Identifiable by Intervening on Retrieval Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02070 - When More Data Is Not Enough: The Context-Sufficiency Frontier in Generative AI Personalization Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00654 - Modeling The Object Representations Underlying Human Physical Reasoning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.12486 - Dependency-Aware Reward Shaping for Agentic Reinforcement Learning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01207 - On Language Drift during RLVR Post-Training Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02015 - PACE: Provenance-Aware Capability Enforcement for Tool-Using LLM Agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01349 - Measuring Iterative Temporal Reasoning with Time Puzzles Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2601.07148 - UrbanVLA: A Vision-Language-Action Model for Urban Micromobility Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2510.23576 - LLM-Assisted Discovery of Typed Semantic Links for Ontology Network Construction Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01393 - SkillLens: Adaptive Multi-Granularity Skill Reuse for Cost-Efficient LLM Agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.08386 - Multicalibration for Unbiased Model-Based Prevalence Estimation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.21549 - Atoms to Processes: The Role of Artificial Intelligence and Machine Learning in Chemical Engineering Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02014 - SIMLIFE: Pattern Understanding for Long-Horizon Human-Agent Partnership Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.19610 - Deny Without Disabling: Authorization-Paired Evaluation and Control for Multi-Agent Systems Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00371 - Ontology-Based Contextual AI Evaluations (OB-CAIE) Methodology Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00529 - Iterative Policy Refinement through Semantic Rollout Analysis Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01652 - Mimir: Physics-Grounded LLM Agents for Long-Horizon Irrigation Control Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02038 - CODesign: Consistency from Data to Trajectory in All-Atom Protein Binder Co-Design Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01773 - Reconstruct, Practice, Go Real: Guided Self-Improvement for Embodied Agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02204 - Geometry-Aware Adaptation for Pretrained Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2307.12226 - Windowed A-K-MDP Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.13676 - Clinical Note Bloat Reduction for Efficient LLM Use Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.16364 - Compute Aligned Training: Optimizing for Test Time Inference Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.24957 - Empty Commitments: When Agents Promise What Their Runtime Cannot Deliver Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01045 - Counterfactual Auditing of Bias in Open-Source Large Language Models for Clinical Triage Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01963 - Auditing Action Settlement in LLM Agent Environments: Order, Progress, and Replay Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01138 - NextMe-800: Anticipating Personal Behavior from Months of Egocentric Video Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01461 - DeepJEPA: Scaling World Models from Within Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00368 - Cog-VADU: A Training-Free Cognitive Reasoning Framework for Video Anomaly Detection and Understanding Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01754 - Query-efficient winner prediction in district-based elections Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00577 - FedLore: Communication and Memory Efficient Federated Learning via Shared Gradient Low-Rank Projection Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01620 - GeoLatent: Geometry-Guided Latent Structuring with Routed Optimization for 3D Reasoning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02091 - A Comparative Explainability Framework for DeBERTa-v3 in Zero-Shot Medical Abstract Classification Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02116 - Easier Said Than Done: Unpacking Intent-Behavior Gap in Jailbreaking LLM-Based Robots Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2412.16633 - Parameter-Efficient Distributionally Robust Adaptation of Tabular Foundation Models under Subpopulation Shift Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01143 - Mem++: Non-Destructive Memory for Long-Term Organizational LLM Agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02002 - The Geometry of Contextual Relations: Language Models Address Facts by Order of Mention Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00910 - The Hitchhikers Guide to Rubric Quality Understanding and Enrichment Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.01375 - PROMO: Preference-conditioned Multi-Objective Reinforcement Learning for Quadrupedal Robots Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01260 - A Simple Doxastic Deontic Logic for Norm-Guided Decision Making Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00668 - WaLLM -- Understanding Use and Engagement with a General-Purpose LLM on WhatsApp Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2505.08894 - Not Too Hard, Not Too Easy: Learning from Intermediate States for LLM Structured Reasoning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.33149 - Pay for the Fault, Not the Flow: Label-Free In-Flow Multi-Agent Workflow Optimization Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01017 - Conflicting Supervision Moves Commitment, Not Capability: A 12.29{\sigma} arrangement effect that is exactly zero under a convention-agnostic score Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00234 - Bridging the Sim-to-Real Gap with multipanda_ros2: A Real-Time ROS2 Framework for Multimanual Systems Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.02269 - Backdoor Purification for LoRA-Tuned LLMs via Null-Space Projection Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00685 - Generalist Representation, Specialist Detection: TS-Router for Time-Series Anomaly Detection Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00978 - DAYJOB: A Benchmark for Long-Horizon Professional Work Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01306 - KaliBench: A Fine-Grained Benchmark for Cybersecurity Tool Use on Kali Linux with Runtime-Free Verifiable Rewards Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02206 - Architecture Without an Architect? Global Governance of Artificial Intelligence in a Divided World Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01716 - Calibration-risk routing for controlled world-model adaptation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01001 - What Does a Sharing Question Add? Auditing LLM Survey Scores for Misinformation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.06820 - Benchmarking Prompt Optimization of Large Language Models With Chess Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00416 - DMAD: Distribution Matching as Adversarial Distillation for Fast Visual Generation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02188 - What Do Rationales Communicate? A Message-Intervention Study in Role-Specialized QA Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00018 - Exposing the Cost of Deep Learning Audio Development Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01619 - FairSSL: Fair Multimodal Self-Supervised Learning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2508.16748 - False Floors: LLM Safety Routing Evaluations Break Under Distribution Shift Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01535 - CoEvolve: Construct-to-Edit Visual Grounding with Bidirectional State Refinement Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01710 - SoftServe: A Scalable Quasi-Newton Method for Deep Learning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02182 - Code That Works, Environments That Don't: Measuring Environment Reproducibility in AI-Generated Software Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00425 - Legal Research Bench: Measuring End-to-End Reliability in Long-Horizon Legal Research Agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00609 - Structured-Noise Masked Modeling for Video, Audio and Beyond Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2503.16311 - Exploring More, Reasoning Better: Stepwise Risk-Sensitive GRPO for Diffusion Language Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00661 - Per-Node Activation Function Evolution in Indirectly Encoded Substrates: Solvability, Limits, and Emergent Diversity Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00149 - Are AI Coders Snitches? An Empirical Study of Pretraining Data Detection on Code Large Language Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2507.17389 - Generation Provenance Before Behavior Attribution: Auditing Synthetic Speech Research Objects Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01378 - Actions with Receipts: Jointly Binding Claims, Evidence, and Execution for Replayable Tool-Agent Auditing Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00327 - Supervising Sound Localization by In-the-wild Egomotion Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01388 - InterviewSim: A Scalable Framework for Interview-Grounded Personality Simulation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.20294 - LLM-based Agentic Reasoning Frameworks: A Survey from Methods to Scenarios Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2508.17692 - Task-Adaptive Grounded 3D-Programmers Using 2D VLMs Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02021 - Why Your Deep Research Agent Fails? On Hallucination Evaluation in Full Research Trajectory Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2601.22984 - Feature Selective Model Collapse in Diffusion Models: Total Replacement versus Fixed-Budget Training Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01318 - DynGhost: Temporally-Modelled Transformer for Dynamic Ghost Imagings Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.10185 - OpenMTB-Audit: Exposing Over-Refusal and Clinical Expert Perspectives in LLM-Based Molecular Tumor Board Safety Evaluation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01497 - A Verifier Can Leak the Answer: Diagnosability Before Optimization in Closed-Loop Agent Debugging Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00126 - Selection-Based Structured Reasoning: Toward Efficient Multimodal Search Agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01892 - A Matched-Budget Audit Framework for Recaptioned Image-Text Supervision Distributions Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00952 - Rethinking Probability-Based Reinforcement Learning From Posterior Concentration Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01458 - Unsupervised Domain Adaptation for Enhanced Radiometer Image Precipitation Estimation using Conditional Flow Matching Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01890 - What Should an Agent Remember? Disentangling Retention from Retrieval in Bounded-Memory Evaluation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00366 - MERID: Multimodal Exploration via Recursive Self-Improvement Agents for Major Depression Analysis Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.36235 - From Order to Distribution: An Exact Operator Framework for Forgetting in Continual Learning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.13460 - Continuous Process-Level Evaluation for Evolving Enterprise AI Agent Skills Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01833 - Pre-training interventions, ex post facto: Grafting model beliefs across checkpoints Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00767 - AiSearch: Interactive Multi-Modal Search with VLMs Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01389 - RelationVGGT: Visual Geometry Transformers for 3D Spatial Relation Segmentation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00970 - Increasing Width Allows Greedy Layer-wise Training to Rival End-to-End Backpropagation in Self-Supervised Learning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00753 - Not All Error Yields to Scale: Where Scaling Stops in Vision-Language Inference Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01640 - TopK-Guided: Adaptive, Budget-Aware Activation Sparsity for Efficient LLM Inference Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01763 - Robust Is Salient: An Informed Adversary Moves the Optimal Signal onto the Salience Pole Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00233 - HHR: Hierarchical Hash Retrieval for Efficient LLM Generation Source: arxiv-cs-ai Topic: AI Research (+AI, Information Retrieval) URL: https://arxiv.org/abs/2610.01230 - Capabilities Ain't All You Need: Measuring Propensities in AI Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.18182 - SCOPE-AD: Sequential cost-aware ordinal-belief planning with energy-based models for diagnostic agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01278 - Outer Diversity of Condorcet Domains Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.00720 - External Observers May See More Clearly: Cross-Model Span-Level Hallucination Detection in Large Language Models via Hidden State Probing Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.02066 - A Multi-Agent LLM Framework for Personalized Health Checkup Interpretation and Guidance Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2610.01451 ## Community - Rex's Dino Store Source: simon-willison Topic: Python (+AI, Django) URL: https://simonwillison.net/2026/Oct/2/rex-s-dino-store/ - Show HN: Made an open-source Lego AI generator Source: hacker-news Topic: General (+AI) URL: https://github.com/anteloc/ldraw-nova - From the creator of Redis; run LLM locally with ds4 Source: hacker-news Topic: General (+AI) URL: https://dwarfstar.sh/ - Rai: CPU-only LLM inference engine in pure Rust Source: hacker-news Topic: General (+AI) URL: https://github.com/Classevelabs/rai - Decision models like Jev don't beat LLM-as-a-judge or traditional classifiers Source: hacker-news Topic: General (+AI) URL: https://developers.redhat.com/articles/2026/10/02/benchmarking-ai-decision-models-against-traditional-guardrails - Greg Kroah-Hartman – Security in the LLM Age [video] Source: hacker-news Topic: General (+AI) URL: https://www.youtube.com/watch?v=NnV_cWeoo5Q - KernelBench: Can LLMs Write GPU Kernels? – Benchmark and Toolkit, Torch –> CUDA Source: hacker-news Topic: General (+AI) URL: https://github.com/ScalingIntelligence/KernelBench ## Media - Build Your Own Open-Source Dots with Pi and Telegram Source: youtube-huggingface Topic: Hugging Face (+AI) URL: https://www.youtube.com/watch?v=HU03WDFB_tQ