TILens Daily Edition 2026-09-02 Filters: topic=ai, github=hidden Stats: 831 articles, 28 sources, 784 research papers Top topics: AI, AI Research, Information Retrieval ## News - Web Search on Amazon Bedrock is now available in AWS GovCloud (US-West) Source: aws-whats-new Topic: AWS (+AI) URL: https://aws.amazon.com/about-aws/whats-new/2026/09/amazon-bedrock-web-aws-govcloud/ - Real-Time Intelligence with IBM Time Series Models on Confluent Source: huggingface-blog Topic: Hugging Face (+AI) URL: https://huggingface.co/blog/ibm-research/real-time-intelligence - ATV Big Air Tour turned 3 days of work into 3 hours with ChatGPT Source: openai-blog Topic: AI Labs (+AI) URL: https://openai.com/index/atv-big-air-tour - Sora API dừng ngày 24/09/2026: checklist cần làm ngay Source: devto-javascript Topic: JavaScript (+AI) URL: https://dev.to/bean_bean/sora-api-dung-ngay-24092026-checklist-can-lam-ngay-57cc - Gemini 3.8 Flash giá $0.75/1M: dev có nên đổi model? Source: devto-javascript Topic: JavaScript (+AI) URL: https://dev.to/bean_bean/gemini-38-flash-gia-0751m-dev-co-nen-doi-model-42dm - How I Built In-Game Hot Reloading with an LLM Chat (Zero Build Steps, Pure ESM) Source: devto-javascript Topic: JavaScript (+AI) URL: https://dev.to/carlos/how-i-built-in-game-hot-reloading-with-an-llm-chat-zero-build-steps-pure-esm-2a2h - 12 Open Source Gems To Become The Ultimate Developer 🔥 Source: devto-javascript Topic: JavaScript (+AI) URL: https://dev.to/anthonymax/12-open-source-gems-to-become-the-ultimate-developer-671 - OpenAI’s new reasoning technique alarms AI safety experts Source: techcrunch Topic: General (+AI) URL: https://techcrunch.com/2026/09/02/openais-new-reasoning-technique-alarms-ai-safety-experts/ - Your next OpenAI API timeout might not be a timeout at all Source: newstack Topic: DevOps (+AI) URL: https://thenewstack.io/astra-api-safety-stops/ - Anthropic’s Claude failures have made agent observability a security priority Source: newstack Topic: DevOps (+AI) URL: https://thenewstack.io/anthropic-claude-agent-security/ - With Gemini 3.8 Flash, Google reminds everyone it's still in the race Source: theregister Topic: General (+AI) URL: https://www.theregister.com/ai-and-ml/2026/09/02/with-gemini-38-flash-google-reminds-everyone-its-still-in-the-race/5294049 - A GeoJSON map viewer Source: adafruit-blog Topic: Hardware (+AI) URL: https://blog.adafruit.com/2026/09/02/a-geojson-map-viewer/ - Google releases Gemini 3.8 Flash, its third Flash model in six weeks Source: ars-technica Topic: General (+AI) URL: https://arstechnica.com/ai/2026/09/google-releases-gemini-3-8-flash-its-third-flash-model-in-six-weeks/ - US government sides with OpenAI on issue of training LLMs on copyrighted material Source: techcrunch Topic: General (+AI) URL: https://techcrunch.com/2026/09/02/u-s-government-sides-with-openai-on-issue-of-training-llms-on-copyrighted-material/ - Google ships its third Gemini Flash model in six weeks Source: newstack Topic: DevOps (+AI) URL: https://thenewstack.io/google-ships-its-third-gemini-flash-model-in-six-weeks/ - 😺 Claude Fable 5.1 can do the work. The hard part is managing it. Source: the-neuron-ai Topic: AI (+AI Research) URL: https://www.theneuron.ai/newsletter/claude-fable-5-1-live-test/ - Why Your RAG System Is Only as Good as Its Translator Model Source: bytebytego Topic: Architecture (+AI) URL: https://blog.bytebytego.com/p/how-to-shrink-a-language-model-without - Anthropic makes changes to stop AI agents running amok, again Source: adafruit-blog Topic: Hardware (+AI) URL: https://blog.adafruit.com/2026/09/02/anthropic-makes-changes-to-stop-ai-agents-running-amok-again/ - Facilitating AI integration with simplicity at scale Source: mit-ai Topic: AI URL: https://www.technologyreview.com/2026/09/02/1142879/facilitating-ai-integration-with-simplicity-at-scale/ - Best AI Presentation Makers for Business in 2026 Source: the-neuron-ai Topic: AI (+AI Research) URL: https://www.theneuron.ai/guides/best-ai-presentation-makers/ - Security updates for Wednesday Source: lwn Topic: Open Source (+AI) URL: https://lwn.net/Articles/1092149/ - Anthropic hires architect of UK AI policy as MPs warn of 'clear conflict of interest' Source: theregister Topic: General (+AI) URL: https://www.theregister.com/public-sector/2026/09/02/anthropic-hires-architect-of-uk-ai-policy-as-mps-warn-of-clear-conflict-of-interest/5293903 - OpenAI Details GPT-Live’s Architecture for Continuous Stateful Voice Interaction Source: infoq-architecture Topic: Architecture (+AI) URL: https://www.infoq.com/news/2026/09/openai-gpt-live/?utm_campaign=infoq_content&utm_source=infoq&utm_medium=feed&utm_term=Architecture+%26+Design - Presentation: Beyond Prompting: Context Engineering for Production-Grade AI Source: infoq-architecture Topic: Architecture (+AI) URL: https://www.infoq.com/presentations/context-engineering-redis-llm-architecture/?utm_campaign=infoq_content&utm_source=infoq&utm_medium=feed&utm_term=Architecture+%26+Design - Fable 5.1 kicks off launch week at the frontier Source: the-rundown-ai Topic: AI (+AI Tools) URL: https://www.therundown.ai/articles/fable-5-1-kicks-off-launch-week-at-the-frontier - 😺Anthropic launched Fable 5.1: and now, the agents cost less Source: the-neuron-ai Topic: AI (+AI Research) URL: https://www.theneuron.ai/newsletter/anthropic-launched-fable-5-1-and-now-the-agents-cost-less/ - Claude Fable 5.1 LIVE: Testing Anthropic’s New AI Agent Source: the-neuron-ai Topic: AI (+AI Research) URL: https://www.theneuron.ai/news/claude-fable-5-1-can-do-the-work-the-hard-part-is-managing-it/ - Everything That Happened in AI Today (Tuesday, September 1, 2026) Source: the-neuron-ai Topic: AI (+AI Research) URL: https://www.theneuron.ai/digest/everything-that-happened-in-ai-today-tuesday-september-1-2026/ ## Research - Avoiding Entity Key Drift in a Data Lake: Step 2, When Fuzzy Matching Stops Working Source: towards-data-science Topic: AI Research (+AI, Python) URL: https://towardsdatascience.com/avoiding-entity-key-drift-in-a-data-lake-step-2-when-fuzzy-matching-stops-working/ - A RAG That Says “Not in This Document” Has to Show Four Kinds of Evidence Source: towards-data-science Topic: AI Research (+AI, Python) URL: https://towardsdatascience.com/a-rag-that-says-not-in-this-document-has-to-show-four-kinds-of-evidence/ - Graph Neural Networks: GCN, MPNN, and GAT, Explained Simply Source: towards-data-science Topic: AI Research (+AI, Python) URL: https://towardsdatascience.com/graph-neural-networks-gcn-mpnn-and-gat-explained-simply/ - A Practical Introduction to PySpark Window Functions Source: towards-data-science Topic: AI Research (+AI, Python) URL: https://towardsdatascience.com/a-practical-introduction-to-pyspark-window-functions/ - Test-Time Scaling in Reasoning LLMs: Inference Regimes, Evaluation, and Reproducibility Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.04001 - A Classifier That Teaches Itself: Self-Improving, Frozen-gate Training (SIFT) for Dynamic Document Classification Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2607.18358 - Convergence issues in Relational Concept Analysis based on AOC-posets Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00054 - Soft-Argmax for the Projective Plane via the Veronese Embedding Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00521 - Online Self-Weighted Fine-Tuning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00734 - EvoFlint: An Evolutionary Atlas of Multi-Turn LLM Vulnerabilities Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00487 - Why Multi-Layer Message Passing Works: Completeness Theory for Graph Neural Network Interatomic Potentials Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00528 - MineDraft: A Framework for Batch Parallel Speculative Decoding Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.18016 - Multi-View Causal Discovery without Non-Gaussianity: Identifiability and Algorithms Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2502.20115 - Right Frame, Wrong Rule: Cultural Cues Expose the Financial Knowledge Gap They Were Meant to Close Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00999 - Process-Aware AI for Rainfall-Runoff Modeling: A Mass-Conserving Neural Framework with Hydrological Process Constraints Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.25093 - Scaled Idempotence in Transformer Attention: Paired OV Geometry and Shared-Value Algebras Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01129 - Semantic-Guided Multimodal Preprocessing for Vision Transformer-Based Clear Cell Renal Cell Carcinoma Grading Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01426 - Efficient Adaptation of ROMs for Unsteady Flows Using Data Assimilation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.23188 - RetroReasoner: A Reasoning LLM for Strategic Retrosynthesis Prediction Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.12666 - FloydNet: A Learning Paradigm for Global Relational Reasoning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2601.19094 - When the Strongest Teacher Is Not the Best Teacher: Student-Centric Answer Selection Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.26872 - GenONet: A Generative operator Network for High-Resolution Precipitation Nowcasting Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00544 - From Truncation to Commitment: Persistent Context in Uniform Discrete Diffusion Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01043 - Independent Reinforcement Learning in Discounted Markov Games Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00504 - Does Fault Localization Beat a Fresh Attempt? A Placebo-Controlled Study of Test-Guided Code Repair Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00854 - Investigating Linear Probe Robustness to Linguistic Register, Medical Specialty, and Corpus Shifts in Medical QA Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01361 - Variance-Adaptive Muon: Pre-Orthogonalization Variance Modulation for Efficient Language Model Pretraining Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2601.14603 - Vision-Language-Guided Pseudo-Labels for Unsupervised Domain Adaptation in Semantic Segmentation for Waste Sorting Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00898 - Breaking the Reasoning Horizon in Entity Alignment Foundation Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2601.21174 - Probabilistic Multi-Agent Aircraft Landing Time Prediction Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2512.08281 - Lingua Franca or Probing Artifact? Rethinking Latent Language in Multilingual LLMs Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00155 - Patterning in Practice: Debiasing Reward Models with Susceptibilities Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00699 - Confess What You Know: Forget-Set Misalignment with Model Knowledge in LLM Unlearning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00605 - TRUST: Threshold-Recalibrated Uncertainty-Safe Training for Certified Dismissal in Breast Cancer Screening Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00300 - Deterministic LLM Inference Across GPU Kernels: Power-of-Two INT8 Quantization Scales and the Limits of Tolerance-Based Conformance Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00363 - A Mathematical Theory of Reusable Neural Bases for Network Compression Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01550 - A convolutional framework for detecting event-driven dynamics in energy price series Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00402 - Bandits in Prod: Hyperparameter Optimization at Inference Time Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01335 - Births are difficult to predict even with rich survey and full-population register data Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01194 - Keep Everyone Happy: Online Fair Division of Numerous Items with Few Copies Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2408.12845 - Exact Risk-Complexity Laws for Projective Boundaries in Scenario Optimization and Distribution-Free Certification Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01355 - BiG-SURE - Bipartite Graph for Semantic Uncertainty and Reliability Estimation of LLMs Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.30646 - DK-GBMKKM: Dynamic Kernel-Space Granular-Ball Multiple Kernel $k$-Means Clustering Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00647 - A Checklist to assess the energy and carbon impacts of ML/AI applications in Earth System Modeling Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00847 - Physiological Information Reliability: Cross-Layer Adaptive Resource Allocation for Cardiovascular Sensing Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00435 - Why Fine-Tuning Encourages Hallucinations and How to Fix It Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.15574 - Explicit Interaction Architectures for Dynamical Learning: A Controlled Study of Structural Inductive Bias Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.19101 - Modelpedia: A Catalog of Model Findings for the Meta-Science of AI Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01090 - Closing the Operational Gap in Semantic Caching Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.19719 - MaskCode: Mask Transformer for Feedback-Assisted Coding With Linear Block Codes Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00715 - Foundation models for electricity price forecasting and battery arbitrage: Can they replace market-specific forecasting models? Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00089 - Self-Reports Are Not Verification: Environment-Grounded Auditing of LLM Operators in Evolutionary Search Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00652 - Validating FKG.in: Soundness Assessment in LLM-Augmented Indian Food Knowledge Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.29249 - MIDR: Enrichment-Augmented Indexing for Multimodal Document Retrieval Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01316 - A Compositional Kernel Model for Feature Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2509.14158 - How Does Alignment Tuning Shape Representations of Sycophancy and Related Cue-Induced Biases in LLMs? Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2607.18114 - S-CEReBrO: Breaking the Memory Barrier in Continuous EEG Monitoring Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2607.27913 - The Constitutional Coverage Trilemma in AI Governance Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01275 - Counterfactual Fragility Certificates: Exposing High-Confidence Brittleness under Structured Evidence Failure Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00366 - Measuring consistency via ensemble margin and local prediction variability: Auditing decision systems in the presence of predictive multiplicity Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01397 - VoiceLongMemEval: Do Assistants Remember How You Sounded? Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00570 - How Do Language Models Choose Between Context and Memory? Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00753 - Channel-Adaptive Edge AI: Maximizing Inference Throughput by Adapting Computational Complexity to Channel States Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.03146 - Different representation learning objectives recover distinct latent structures from the same psychometric data Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00100 - Recent Developments in Transformer Inference Deployment on FPGA Platforms: A Survey Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01212 - Common-Center Geometry and Certified Radial Reconstruction for Energy-Form Full Conformal Regions Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.24964 - SEBA: Sample-Efficient Black-Box Attacks on Visual Reinforcement Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2511.09681 - Group Adaptive Clipping Policy Optimization Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00444 - mzCache: On-Device LLM Memory Management under Multitasking Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01338 - Hidden relationships in a document-derived property graph: top-k chunk embeddings and inverse-distance weighting over a dynamically evolving ontology Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00387 - Performance-Efficiency Tradeoffs in Transformers: An Approximation Theory Perspective Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2510.03784 - Dr. Claw: An AI Scientist Workspace for Vibe Research Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00365 - Persistent Entropy as a Detector of Phase Transitions Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.09058 - Poisson-Gamma Dynamical Systems with Time-varying Transition Dynamics Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00896 - Stochastic complexity of vectors containing cluster structure Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00084 - Replicating TRACE: A Practitioner's Guide to Its Threshold and Particle Budget Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01108 - Neural Symbollic Regression Using Deep Learning and Sparse Modelling Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01102 - Ontology-Guided Neuro-Symbolic Inference: Grounding Language Models with Mathematical Domain Knowledge Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.17826 - Generalization Bounds for Markov Algorithms through Entropy Flow Computations Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2502.07584 - The Structure of Quantization Damage in LLMs: Why the Next Bit Should Be Spent Globally Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01587 - Exact Global MCMC with Denoising Diffusion Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00279 - The Frame Kernel Method for Multiscale Operator Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.25084 - Compositional Machine Design as Program Synthesis with LLMs Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2510.14980 - Subliminal Learning as Trait-Direction Drift: A Mechanism and Targeted Control under SFT Distillation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01091 - Polarizable atomic multipoles for learning long-range electrostatics Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.05746 - Risk-Aware Decision-Making for Autonomous Overtaking: A World Model-Based Mixture-of-Experts Framework Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00385 - HarnessEvolve: Learning from Reference Trajectories for Reliable Agent Self-Evolution Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00829 - Are You Thinking What I am Thinking? : Examining Conceptual Separation in Neural Architectures Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00764 - SAGE: Subpopulation-Aware Generative Enhancement for Mitigating Spurious Correlations Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01051 - Subspace Levenberg Marquardt Algorithms in Training Neural Networks Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00789 - SAGE: State-Grounded, Abstention-Aware Evaluation of Task-Oriented Dialogue Agents Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00434 - BiasGym: A Simple and Generalizable Framework for Analyzing and Removing Biases through Injection Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2508.08855 - NashDreamer: Model-Based Reinforcement Learning for Zero-Sum Imperfect-Information Games Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01549 - Beyond Dense Adam States: Adaptive Log-Space Quantization for Memory-Efficient Optimizers Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.22322 - A Study of Hidden-State Optimization Order in Predictive Coding Networks Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00686 - HarmoCore: Functional Latent Diffusion for Sparse Reconstruction of Oscillatory Wave Fields Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00679 - Control Variate Score Matching for Diffusion Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2512.20003 - Learning to Refine Hidden States for Reliable LLM Reasoning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.17524 - ToSCA: Leveraging Hierarchical Reinforcement Learning on Temporal and Strategic Abstractions of Conversational Agents Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.21969 - Integrated Noise and Safety Management in UAM via A Unified Reinforcement Learning Framework Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2508.16440 - Lightweight Adaptation of EEG Foundation Models for Stroke Motor Imagery Decoding: Domain Shift and Subject-Level Robustness Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00282 - D3-Gym: Constructing Real-World Verifiable Environments for Data-Driven Discovery Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.27977 - Breaking the Structural Identity: Personalized Federated LoRA Fine-tuning under Rank Heterogeneity Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00632 - On the Reliability of Generative Augmentation: A Wasserstein-Based Theoretical and Empirical Study Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01410 - DOG-DPO:Dynamic Optimization in Geometry for Safety Alignment Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.07678 - What Drives Representation Steering? A Mechanistic Case Study on Steering Refusal Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.08524 - On the Existence of Consistent Adversarial Attacks in High-Dimensional Linear Classification Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2506.12454 - SCALE:Scalable Conditional Atlas-Level Endpoint transport for virtual cell perturbation prediction Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.17380 - Fair Minimum Labeling: Efficient Temporal Network Activations for Reachability and Equity Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2510.03899 - Field-Aware Agent Skill Retrieval Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.02880 - Manifold-Aware General Coded Computing for Straggler-Resilient Distributed Computing Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00552 - 3D-Consistent Multi-View Editing by Correspondence Guidance Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2511.22228 - Relational Task Generation Language: A Declarative Specification Framework for Relational Deep Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01292 - GazeRefine: Expert Gaze as a Test-Time Prompt for Training-Free Medical Image Segmentation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01310 - A hybrid quantum-classical neural network for learning to route Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00489 - Sharp Mixed Spectral Barron Regularity of Coulombic Many-Electron Wave Functions Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00872 - Accelerating Reinforcement Learning via MPC Solver-Gradient Guidance for Weights-varying MPC Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01061 - Does Imitation Learning Preserve Temporal Robustness in Dexterous Manipulation? An Expert-Learner Comparison Across Task Execution Speeds Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01453 - Why Do Reasoning Models Lose Coverage? The Role of Data and Forks in the Road Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.17026 - Optimizing Byzantine Node Placement in Decentralized Federated Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01495 - OCGQuant: Outlier-Companion Grouping for NVFP4 Quantization Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00066 - Unsupervised Partner Design Enables Robust Ad-hoc Teamwork Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2508.06336 - FractalNet-Based Heterogeneous Federated Learning for Orbital Edge Intelligence in Satellite Mega-Constellations: A Wildfire Case Study Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00875 - TopoCompress: Long Context Compression via Graph-Wired Semantic Trajectories Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.30811 - Neurosymbolics for Data Engineering: Achieving Long Context Token Reduction Without Finetuning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00367 - Do LLMs Know Your Neighborhood? Auditing LLM Priors for Neighborhood-Level Mobility Prediction and Structural Alignment Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00345 - Inverse Reconstruction of Shock Time Series from Shock Response Spectrum Curves using Machine Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.03229 - Safin-1: Safety from Within through Memory-Native State Evolution Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00092 - RECAP: Regression Evaluation for Continual Adaptation of Prompts Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.06698 - GENIE: Watermarking Graph Neural Networks for Link Prediction Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2406.04805 - I-CARE: Analysis of interference-related phenomena in a controllable, diverse and representative unlearning setting for text-to-image models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00003 - EM^2Mem: Event-Centric Multimodal Memory for Large Language Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00551 - ES-AHD: An Evolution Strategy Framework for Automatic Heuristic Design Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00023 - Logarithmic-Free Moment and Generalization Bounds for Uniformly Stable Algorithms Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.09870 - Where the Verifier Fails: A Category-Level Audit of Reward Signals in RLVR Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01354 - UI-Venus-2 Technical Report Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00028 - Provably Efficient Federated Reinforcement Learning with Linear Function Approximation and Logarithmic Communication Cost Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00193 - Geometry-aware Latent Autoregressive Generative Model for PDEs in Complex Domains Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00297 - Efficient Learning of Balanced Signed Graphs via Sparse Linear Programming Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2506.01826 - Iterative GRPO: Batch-Online Policy Iteration for Multi-Turn RL via Single-Turn RLHF Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2511.21638 - HiMPO: Hindsight-Informed Memory Policy Optimization for Less-Entangled Credit in Long-Horizon Agents Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.16285 - Efficiently Estimating Optimal Hyperparameter Scaling Laws through Power-Law Entropy Search Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01431 - Learning Sparse Decision Trees via Transformer Variational Auto-Encoders Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01430 - Dense Weak Hiding: Closing Complexity Gaps in Nonconvex and PL Finite-Sum Optimization under Individual Smoothness Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00045 - Reading the Gate, Not the Interference: Output-Side Interference Measurement Does Not Track Merge Collapse Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.11797 - Faster Than Flash: Exploiting Attention Sparsity for Efficient Long-Context Decoding Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00097 - Direct Optimization of a 3D Finite-Source Reflector via Neural-Network Parameterization Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00899 - The Visual Insensitivity Gap: Diagnosing When Vision-Language Models Fail to Use Visual Evidence Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00868 - Dense Process Supervision for Search Agents via Fact Utility Estimation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00833 - SinkPruner: Sink-Free Visual Token Pruning for Multimodal Large Language Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01004 - Provably Safe Sim-to-Real Transfer Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01418 - EchoSonar-R: A Multi-View Reasoning-Enabled Model for Disease Classification and Report Generation in Echocardiography Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.28164 - The Alexander-Hirschowitz theorem for neurovarieties Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2511.19703 - MUGEN: Generating Unlearnable Graph Examples for Multiple Learning Tasks Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00696 - GeoPAR: Large-Scale Multi-Agent Combinatorial Optimization with Geometry-Guided Parallel Autoregressive Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00577 - Towards unsupervised representation learning for quantum data: quantum models with inference and generation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00372 - SupraTok: Cross-Boundary Tokenization for Enhanced Language Model Performance Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2508.11857 - TACS: Trajectory-Aware Candidate Selection for LLM Jailbreak Suffix Optimization Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.29564 - The Curse of Multilinguality in Lexical Normalization Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00329 - Scaling Near-Optimal SFT-RL Annotation Budget Allocation from Small to Large LLMs Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01573 - DiscoverPhysics: Benchmarking LLMs for Out-of-the-Box Scientific Thinking Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.26087 - Towards Agentic Cloud Engineering: Graph and Loop Engineering with a Zero-Trust Agent Harness Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00050 - A Stable Aggregation Method for Quantum Federated Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00356 - Beyond AHI: An Interpretable Causal-Discovery-Guided Framework for Sleep Recovery in Connected Health Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.18506 - FedSPDnet: Geometry-Aware Federated Deep Learning with SPDnet Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.22494 - EEG-VID: Task-Guided Latent Predictive Pretraining for EEG Decoding and Assistive Target Selection Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00566 - NeuroPriv: Adversarial Representation Learning for Privacy in Wearable EEG Systems Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00390 - KV Cache Offloading for Context-Intensive Tasks Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.08426 - Nonlinear Dynamics In Optimization Landscape of Shallow Neural Networks with Tunable Leaky ReLU Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2510.25060 - ValueGraph: Value-Signal Guided Graph Pre-training for Contextualized User Representation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00057 - Accelerating Chemical Kinetics for Exoplanet Atmospheres using Neural Networks Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00428 - DualStake: Dual-Path Confidence Calibration in Deep Research Agents Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00935 - Text Capability Loss in Vision-Language Adaptation: An Attention-Sink Diagnosis Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00746 - QABBA: Symbolic Time-Series Compression via Integer-Quantized Aggregation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2411.15209 - Are Near-Tied LLM Rankings Robust to Family-DIF-Guided Benchmark Recomposition? Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00482 - Global Attention with Linear Complexity for Exascale Generative Data Assimilation in Earth System Prediction Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.16590 - SOVER: Formal Certification of Optimization Reformulations via LLM-Assisted SMT Verification Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00728 - EEG-AS: Instance-Level Foundation Model Selection for EEG Foundation Models via Behavior Reconstruction Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00653 - Fractal dimension predicts quantum kernel collapse in angle-encoded data Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00475 - Edge-Girth as a Structural Edge Feature for Graph Neural Networks Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01441 - Variable Selection for Feature-Based Newsvendor Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01544 - Conditional Flow Matching for ML-Based Inverse Design Problems Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00863 - Final Checkpoints Are Not Enough: Analyzing Latent Reasoning Faithfulness Along Training Trajectories Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2607.06648 - Multi-Step Knowledge Interaction Analysis via Rank-2 Subspace Disentanglement Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2511.01706 - REFACTOR-VLA: Unsupervised Library Learning of Typed Motor Programs Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01215 - Matched Queries for Curvature and Density at Branching Junctions Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01319 - Elite-Weighted Supervised Fine-tuning for Goal-Directed Molecular Optimization Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00189 - WHALE: A Simple Recipe for Joint Harness-Weight Optimization Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00196 - Artificial Rosetta Stone: Constrained Maximum A Posteriori (MAP) Reconstruction of Symbolic Raga Sequences via Order-k Markov Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01064 - Any-Order GPT as Masked Diffusion Model: Decoupling Formulation and Architecture Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2506.19935 - Debiased Inference for AI-Generated Data without Gold-Standard Labels: Identification via Multiple Imperfect Measurements Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.18294 - Backspace as a Natural Experiment: An Accelerated Failure Time Model of Selective Post-Error Motor Impairment in Parkinsons Disease Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2607.24796 - Neural means and kernel corrections for operator learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00389 - Is Knowledge Distillation Actually Greener? A Case Study in Machine Translation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.09691 - Steer, Don't Solve: Training Small Critic Models for Large Code Agents Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.21811 - Enabling KV Caching of Shared Prefix for Diffusion Language Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.07571 - CRAD: Class-wise Reliability-Aware Distillation for Decentralized Heterogeneous Federated Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00446 - SMELT: Scaling Laws for Compute-Matched MoE Looped Transformers Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01343 - Local Reference Geometry Residual Augmentation for Imbalanced Time Series Classification Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00093 - Verdict Instability of OOD Scores under Reference Resampling Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00691 - From Language to Behavior: Scaling Sequence Transformers for Industrial Recommendation Ranking with Rec-Native Designs Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01240 - Leakage-Audited Benchmarking Reveals Limited Evidence for Cross-Subject Auditory-Evoked EEG Vowel Perception Decoding Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.00865 - Facet-0: A Robotic Foundation Model for Contact-Rich Precise Manipulation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01596 - When Metropolis and Hastings Meet Bradley and Terry: Exact MCMC From Preference Voting Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00905 - Explore More, Drift Less: Outcome-Only Reinforcement Learning Can Suffice for Long-Horizon Interactive Agents Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01245 - Beyond Scores: Understanding LLM-as-a-Judge Mechanisms in Summarization Evaluation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01604 - WiSDoM: Wireless Sparse Decision Transformer with Mixture-of-Experts for Multi-Task Mobile Network Optimization Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00284 - Agentic Empirical Asset Pricing: Methodological Foundations Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00731 - CompanionSim: Synthetic Data for Evaluating Anthropomorphism in Human-AI Relationships Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00250 - Can LLMs Discover Scientific Laws in Real and Parallel Worlds? Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01552 - Task-Specific Prompt with Global Context for Multi-Task Graph Pre-Training Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00047 - Adapting Without Gradients: Affine Statistics Transport and What Its Certificate Can Tell You Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00374 - Selective Agent Guidance via Entropy: Learning Autonomous Policies from Imperfect VLM Teachers Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01567 - Capability-Gated Language Models: Security Composes, Utility Does Not Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00445 - Embedded Conditional Independence Tests for Large Language Model Generated Text with an Application to German Parliament Speeches Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00946 - Hyper-Fold: Exploring the Expressive Limit of Sequence-Geometry Learning for Proteins via Hypergraph Modeling Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.29207 - QTEA: Ternary LLMs with Sparse Residual Salient Weight and By-Column Optimization Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00224 - Good Memory Has ECC: Evaluating the Memory of Vision-Language Models Beyond Accuracy Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00103 - Building Expressive and Tractable Probabilistic Generative Models: A Review Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2402.00759 - Sierpi\'nski--Knopp Wasserstein Distance for Persistence Diagrams and Applications to 2-Wasserstein Approximation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01528 - Context-Grounding Gains Are Mediated by Pre-existing Machinery: Auditing GRPO, SFT, and DPO Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00925 - Position: Privacy Is a Claim, Not a Property of Synthetic Data Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01273 - Topic Matching in the Wild: Benchmark and Lessons from Real-World ASR Transcripts Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00330 - BeamRMX: Radiation-Pattern-Driven Learning for Generalizable Beam Radio Map Prediction and Beam Management Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00615 - HEAL: Hindsight Entropy-Assisted Learning for Reasoning Distillation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.10359 - Learning Task-Specific Antibody Representations via Function-Aware Masking Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00518 - FlexP-SFT: A Flexible Aggregation-Free Framework for On-Device Personalized Split Federated Fine-Tuning of LLMs Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2508.10349 - The Topological Trouble With Transformers Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.17121 - Freeze, Diffuse, Decode: Geometry-Aware Adaptation of Pretrained Transformer Embeddings for Antimicrobial Peptide Design Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2511.23120 - Rethinking Learnability in Offline Data-driven Optimization Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01493 - Pre-carved Niches: The Formation Dynamics of Modular Task Partitions in Early LLM Training Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01170 - Frozen Cores Need Task Signal: Fisher-Whitened Cross-Covariance for Low-Resource LLM Adaptation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00762 - Workload Identification with Physical Side Channels for AI Governance Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00309 - When Prediction Error Is Not Enough: Evaluating Nuisance-Function Prediction for Causal Estimation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00071 - FedReview: Review and Dispose Poisoned Updates without Validation Datasets or Historic Knowledge Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2402.16934 - Diffusion as a Training Curriculum for Timestep-Free Iterative Reasoning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01449 - Contribution-Aware Bandwidth Allocation for Multimodal Split Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01406 - Assessing Alignment and Stability of Feature Importance Explanations via Weight of Evidence Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00090 - Hidden State Poisoning Attacks against Mamba-based Language Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2601.01972 - Cheap Verifiers, Large Blind Spots: Measuring the Reliability Cost of Cost-Saving Cascades Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01345 - What Do Students Learn? A Feature-Level Analysis of Dark Knowledge Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.03052 - Skill Reuse as Compression in Agentic RL Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.31509 - LatentPress: Context Compression Beyond Text and Vision Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01507 - Auditing Frozen-Encoder Anomaly Detection Across Mechanical Systems: Representation Provenance, Calibration, and Protocol Effects Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2601.11415 - Flawed in Nature, Perfect through Evolution Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00129 - Rigorous Error Certification for Neural PDE Solvers: From Empirical Residuals to Solution Guarantees Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.19165 - Universal Approximation of Nonlinear Operators and Their Derivatives Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.15285 - LLM-driven design of physics-constrained constitutive models: two agents are better than one Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.23754 - When Does Online Adaptation Pay on the Edge? A Leakage-Free Evaluation of Warmup, Learning-Rate Selection, and Resource Trade-offs for Time-Series Forecasting Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01126 - CRAFT: Fine-Tuning Pre-hoc Explainability in AI-native 6G RAN Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00590 - HBQ: Hierarchical Scaling Block Quantization with Hardware-Efficiency-Aware Design for Accurate LLM Inference Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00450 - Superposed Latent Autoencoder Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01158 - Deep learning based numerical approximation algorithms for stochastic partial differential equations Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2012.01194 - Solving In-Table Prediction Problems by Deep Neural Networks with Performance Evaluation Using Synthetic Data Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01262 - Towards Provable and Scalable Training of Quantized Neural Networks with Ising Optimization Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2506.18240 - One-Layer Transformer Provably Learns Multiclass One-Nearest Neighbor in Context Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01311 - Advantage Weighted Matching: Aligning RL with Pretraining in Diffusion Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2509.25050 - VATO: A Vortex-Force-Aware Transformer Operator for Unsteady Separated Aerofoil Flows Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00507 - Training-Free Policy Violation Detection via Activation-Space Whitening in LLMs Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2512.03994 - Can LLMs Imagine Moral Alternatives Beyond Binary Dilemmas? Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.31213 - Predicting Subsurface Abnormalities Growth using Physics-Informed Neural Networks Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01417 - Topological Steering Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00597 - Modeling Information Blackouts in Missing Not-At-Random Time Series Data Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2601.01480 - Can LLMs Use Relational Transformer Embeddings? Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00457 - Learning to Remember: End-to-End Training of Memory Agents for Long-Context Reasoning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.18493 - PEARL: Training Socratic Tutors with Pedagogically Aligned Reinforcement Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.29582 - Online Regime-aware Calibration for Black-box Social Simulators via Posterior-assisted Evolutionary Dynamic Optimization Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2601.19481 - ReNFT: Repairing Mode Collapse in Reward Post-Training via Internal Probability-Mass Recalibration Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00061 - How Temporal Correlations Shape Memory in Linear Recurrent Neural Networks Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00420 - Disciplined Bilevel Programming Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00644 - Exploring Sparse Autoencoders in Text-Based Causal Confounding Adjustment Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01322 - Denoising Diffusion Generative Models Secretly Calculate Attentions Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00885 - A Multi-Branch Feature Fusion Approach for Health Misinformation Detection and Propagation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00403 - Coordinate-Residual Physics-Driven Neural Network for Inverse Scattering Imaging Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.09382 - Real-Time Neuromorphic Spectrum Intelligence Simulator Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00585 - Web Price Extraction: State of the Art and an Adaptive Browserless Implementation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01030 - Silence is Golden: Mitigating Hallucinations in Large Audio-Language Models via Layer-Weighted Vector Steering Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2510.12851 - Post-Training Science for Supervised Fine-Tuning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01244 - Denoising the Deep Sky: Physics-Based CCD Noise Formation for Astronomical Imaging Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2601.23276 - TRIAGE: Three-level Routing and Intelligent Agent Guidance for Efficient Execution Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01428 - The Multiple Timescales of Gradient Descent on the Edge of Stability: A Perturbative Derivation of the Central Flow Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01034 - CopyShield: A Cross-Level Benchmark of Copyright Defenses in LLMs Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01161 - Uniform a priori bounds and error analysis for the Adam stochastic gradient descent optimization method Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.18899 - DynaNDE: Dynamic Near-Data Expert Scheduling for Batched MoE Inference Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00407 - Context Window Failures in Relational Foundation Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00460 - Stress Testing Unlearning Algorithms Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.22527 - CATeye: Coupled Attribute-Topology Invariance Learning for Voucher Abuse Detection Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01425 - Attention Sensitivity Is Not Enough: Dissociating Attention-Level and Behavioural In-Context Learning under Fine-Tuning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00064 - Let Confidence Change, Not the Prediction: Prediction-Preserving Repair for Post-hoc Calibration Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01072 - MMAI Gym for Science: Training Liquid Foundation Models for Drug Discovery Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.03517 - A penalised Saito functional for heuristic search of free line arrangements Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.02995 - CADKnitter: Compositional CAD Generation from Text and Geometry Guidance Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2512.11199 - Shallower ReLU Network Representations via Exact Linear Algebra Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2607.21651 - DAGGER: Distractor-Aware Graph Generation for Executable Reasoning in Math Problems Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2601.06853 - Can machines think efficiently? Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2510.26954 - Three-dimensional Conditional Diffusion Models for Cosmological 21 cm Lightcone Emulation Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.29016 - Higher Structures in Deep Learning Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00472 - Prediction-Assisted Pricing and Admission for LLM APIs with Stochastic Token Consumption Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00710 - MRMAD: A Multi-Round Multi-Audio Benchmark for Evaluating Acoustic Degradation Perception in Large Audio-Language Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.22236 - AutoScientist-Quant: Self-Evolving Coding Agents for Automatic Research in Quantitative Investment Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.28632 - Latent Recurrent Transformer: Architecture Exploration, Training Strategies, and Scaling Behavior Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.26797 - Controllable Image Captioning with Prompt-Conditioned Scene Rewards Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00709 - Quantum Sparse Autoencoders for Q-Matrix Estimation in Cognitive Diagnosis Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01537 - Online simultaneous inference for quantiles via smoothed stochastic gradient descent Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2505.13299 - ViTAMINS: An Empirical Study of Training Self-Supervised Vision Transformers with Synthetic Hard Negatives Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01041 - Latent-Space No-Arbitrage Geometry of Generative Models for Implied Volatility Surfaces Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00332 - Non-Parametric Spatiotemporal Trajectory Prediction via State-Conditioned Transition Sampling Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.14349 - Synthetic Worlds for Temporal Evaluation and Knowledge Updating in LLMs Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00184 - DeSyR: A Decoupled Symbolic Recovery Framework with PINN-Guided Structure Search and Physics-Informed Coefficient Refinement Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00530 - Gradient-Update Mismatch: Rethinking Conflict-Free Training of Physics-Informed Neural Networks Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01558 - RW-LoRA: Communication-Efficient Decentralized LoRA Fine-Tuning via Random Walks Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00078 - AdaptNTK: Adaptive Uncertainty Quantification and Active Learning for Neural Network Potentials Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00488 - Performative Privacy: When Differential Privacy Maximizes Utility Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.28198 - ReproRepo: Scaling Reproducibility Audits with GitHub Repository Issues Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.18237 - Spawn Freely, Act Sparingly: Progressive Risk Vesting for Recursive LLM-Agent Trees Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01035 - Retrieved but not ranked: surface-form bias in structural retrieval, from mathematics to agent trajectories Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01556 - Recurrent State Encoders for Efficient Neural Combinatorial Optimization Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2509.05084 - Can We Trust In-Distribution Success? Locked Evaluation Reveals Transfer Failure and Sampling-Depth Entanglement in CRISPRi Perturbation Prediction Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.00152 - REAL-Q: E2E LLM Quantization via Dynamic Gradient Descent Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00049 - MemoryWalker: Stop Training Agents on Contexts They Never Saw Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00865 - Reparameterization through Coverings and Topological Weight Priors Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.23804 - Semi-Supervised Classification with Informative Missing Labels in Weibull Mixture Models Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00774 - iPINN for Broadband CARS Phase Retrieval: A Framework for Function Approximation and Inverse Modeling Problems in Nonlinear Spectroscopy Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00883 - Multi-Head Self Attention is a Parameter Identification Mechanism Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01231 - AgentProv: Auditing Agentic LLM API Providers via Tool-use Policy Probes Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00052 - Generative artificial intelligence for reliable mechanistic reasoning for corrosion Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00099 - DISTAL: Distillation and Self-Supervised Pretraining for Structure-Agnostic Materials Property Prediction Source: arxiv-cs-lg Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00059 - From Prompt to Purchase: How AI Brand Recommendations Move Consumers on the Open Web Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2606.10907 - REIGN: Refurbished Embeddings with Integrated Guidance Networks for Efficient Context-Length Scaling Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2608.29899 - NeuroGraph: An AI Graph-Driven Neuro-Symbolic Framework for Explainable Threat Reasoning in Advanced Manufacturing Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2609.00604 - TRIS: A Tri-Layer Retrieval Integrity Sieve Against Knowledge Poisoning Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2609.00470 - SilentProbe: Measuring Silent Failure in Production APIs Used as Agent Tools Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2609.00035 - Sources of Truth: A Multi-Platform, Multilingual Audit of Citations in AI Mental Health Information Queries Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2609.00319 - AutoConcept: Training-Free Concept-Guided Reranking for Metadata-Available Composed Image Retrieval Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2609.01456 - TGR: Advancing Industrial Recommendation from Generative-Paradigm Ranking toward Unified Generation and Reasoning Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2609.00986 - RATIO: A Benchmark for Retrieval Across Typed Ideation Operations in Scientific Literature Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2608.27394 - From Saliency to Discriminability: Rank-Preserving Visual Token Pruning for VLM Rerankers Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2609.00667 - TC-RAG:Turing-Complete RAG's Case study on Medical LLM Systems Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2408.09199 - E-SENS: Exclusion-Sensitive Penalization for Negative-Constraint Retrieval Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2608.30130 - MUSES: A Benchmark for Prospective Intellectual-Roots Retrieval Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2609.00313 - Ctrl-F-Resist. Practices, Challenges, and Technical Needs of Civil Society Organizations Monitoring the Far-Right Online Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2609.00808 - SwapRec: Warming Up Cold Items Through Training-Time Swaps Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2609.00913 - VerTox: Verifiable Reward-Guided Corpus Poisoning Against Neural Ranking Models Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2609.01325 - Closed Forms and Synthetic Twins: Predicting Approximate Nearest Neighbor Recall from Embedding Statistics Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2609.00364 - It Takes Two to Match: Co-Evolving Generative Retriever with Reinforcement Learning Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2609.00638 - DCA-MoE: Spatially Adaptive Cross-Layer Fusion and Density-Routed Experts for Crowd Counting Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2608.15213 - Towards Effective Structured Context Modeling for Conversational Recommender Systems via Dual-node Monte Carlo Tree Search Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2609.00618 - World Model-Guided Reinforcement Learning via Counterfactual User Engagement Simulation Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2609.01067 - Two-Sided State-Space Models for Sequential Recommendation with Non-Random Multimodal Review Feedback Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2609.00165 - APEX-EM: Non-Parametric Online Learning for Autonomous Agents via Structured Procedural-Episodic Experience Replay Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2603.29093 - Staged Linguistic Seeding: Grounded Query Expansion for Verified-Unit QA in AI Contact Centers Source: arxiv-cs-ir Topic: Information Retrieval (+AI, AI Research) URL: https://arxiv.org/abs/2609.00844 - Auditing Harness Tampering in Self-Improving Agents Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00069 - A Certificate-Producing Cascade for Equational Implication: The SAIR EQT2 Stage 2 Solver Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00706 - Visual Framing for News Stance Detection via Image Generation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00685 - RePro: Proof-Verified Benchmark Rewriting for Reliable Evaluation of LLM Mathematical Problem Solving Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00062 - Lazy Grounding: Attacking Search Agents with Factual Evidence Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.30303 - Uncovering and Mitigating Aggregation-Induced Reward Hacking in Multi-Reward Reinforcement Learning Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00213 - CyberFactory: Scaling Cyber Security Capabilities with Instances from the Wild Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.23181 - Behaviorally Grounded User Profiles from the Wild for Personalized Alignment and Multi-Perspective Reasoning Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00014 - Behaviorally Effective LoRA Writes Are Sparse and Structured Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01374 - HiVe: Beyond Static Prompts for Multitask Learning via Hierarchy-based Vertical Mixture-of-Experts Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.29790 - IntroConformal: Conformal Factuality Guarantees for Large Vision-Language Models via Introspective Signals Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01375 - InSight: A Benchmark for Agentic Claim Verification in Interactive Visualizations Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01383 - The Rise of Verbal Reinforcement Learning Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01597 - NSIDDx: A Design Framework for Neuro-Symbolic, Practitioner-First Differential Diagnosis in Low-Resource Settings Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00256 - SWE-bench Science: Can Coding Agents Resolve Engineering Tasks in Science? Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.19799 - AI Alignment through a Game-theoretic Lens: A Survey Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.27910 - Polished but Unresolved: Identifying Late-Stage Pressure States in Long-Horizon Tool-Use Agents Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00823 - From Rollouts to Recipes: Self-Contained Post-Training for LLMs Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01422 - From Confusion to Clarity: Confusion-Aware Retrieval and Knowledge Injection for Text Classification Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01564 - When Compression Helps and When It Hurts: Condition-Aware Analysis of Chain-of-Thought Distillation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.21704 - Automated Researchers Can Reliably Mitigate Alignment Failures Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.28945 - Will the User Ever Know? Covert Indirect Prompt Injection Attacks on Tool-Using LLM Agents Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.30362 - Anamnesis: An Open-Source Platform for Large-Scale Backstory-Conditioned Survey Simulation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2607.10628 - Assessing Suicide Risk in Arabic Crisis Helpline Calls: A Comparison of Arabic and English Large Language Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00191 - From Production Traffic to Post-Training: Building a Self-Hosted LLM That Covers the Corporate Request Mix Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01572 - OUTLETS: Output-Length Prediction from Speculative Decoding Backbones Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01068 - StateSwap: Probing Support-Elimination Hidden States in Multiple-Choice Questions Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01081 - When Irrelevant Text Matters: Affine Margin Shifts in Multimodal Large Language Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.19208 - Value Over Language Model: Detecting Original Contribution in Writing Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00700 - Suffix-Constrained Greedy Search Algorithms for Causal Language Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.01243 - Bridging Lexical Divergence: LLM-Assisted, Cost-Efficient, Zero-shot Scientific Entity Linking Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00228 - Lagged Coupling: Internal Representations Become Readable Before They Become Causal Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01048 - CoMMET: A Psychologically Grounded Benchmark for Evaluating Theory of Mind in Multimodal LLMs Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.11915 - Trust Your Guide Only When Certain: Uncertainty-Aware Sparse Alignment at Inference Time Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00624 - Joint Training Is Not Enough: Conditioned Cross-Granularity Training for Multimodal Document Understanding Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00756 - Calibrating Small Language Models for Claim Check-Worthiness Detection Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.30731 - Quit While You're Ahead: Quit for Efficient Candidate Generation in Machine Translation Reranking Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00588 - HugAgent: A Human Simulation Benchmark for Individual-Level Reasoning Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2510.15144 - LLMPEDIA: Browsing, Verifying, and Comparing the Parametric Encyclopedic Knowledge of LLMs Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01182 - LLM-Driven Autonomous Vehicles Inherit Human Driver Biases in Pedestrian Yielding: Results and Implications From A New Benchmark Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00192 - Location-Aware Language Models via Secondary Embeddings Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00454 - GRRM: Group Relative Reward Modeling for Machine Translation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.14028 - Late Transformer Layers Recode Syntax Canonically: Evidence from Greek Scrambling and Cross-Layer Generalisation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00416 - DiscoTrace: Representing and Comparing Answering Strategies of Humans and LLMs in Information-Seeking Question Answering Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.15140 - Apples on the Table? Evaluating Text-Guided 3D Scene Synthesis via Fine-Grained Constraint Verification Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2511.03001 - Toward Workflow-Aware Benchmarking for Healthcare NLP Agents Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00296 - Do General NLP Embeddings Capture Ontological Reasoning? Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00177 - Reliability Challenges in Diffusion Vision-Language Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01318 - Citing Less Critically: LLMs Reshape the Rhetoric and Reach of Scientific Citation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01432 - Aligned but Flattened: Analyzing the Trade-off between Cultural Alignment and Diversity in LLMs Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00565 - Towards AI-Assisted Clinical Trial Matching: Practical Considerations, Multicenter Evaluation, and Real-World Deployment Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01202 - CUDA-Harness: Harnessing Agentic CUDA Kernel Generation and Optimization from Natural Language Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00058 - Removable and Irreducible: A Token-Cost Ledger for the Multilingual Tokenization Tax Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00378 - DART: Draft-Agreement Routing for Training-Free Adaptive Thinking Budgets in Hybrid Reasoning Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.23181 - When History Is Multimodal: Rethinking Context Management for Long-Horizon Agents Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.29897 - Predicting Program Exit Code with LLMs and Programming Language Semantics Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00579 - PromptNCE: Conditional Probabilities and PMI Using Only LLMs and Contrastive Estimation Prompts Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.21776 - Closing Cost-Quality Gap in Document VLMs: Difficulty-Aware Data Curation and Quality-Adjusted Deployment Economics Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01575 - Can LLMs Reliably Self-Report Adversarial Prefills, and How? Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.23671 - DECSELFMASK: Leveraging Unlabeled Text via Self-Relevance-Guided Masking for Decoder-Only Classification Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.09466 - Zero Hallucination, by Construction: Hallucination-Aware Layered Oversight for Trustworthy Enterprise AI Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2607.17883 - When Tokenization is Secretly Output Supervision Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01386 - Unified Multi-Dialectal Neural Machine Translation for Bangla Using the Dwadash Benchmark Corpus Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.12018 - How Human-Like Are Large Language Models? A Register-Aware Linguistic Evaluation Framework Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.23651 - CWoMP: Morpheme Representation Learning for Interlinear Glossing Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.18184 - Camellia: Benchmarking Cultural Biases in LLMs for Asian Languages Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2510.05291 - (V)LMs generalize beyond surface co-occurrence: Evidence from cross-modal number agreement Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00443 - Scientific Agent Skills: A Library of Procedural Knowledge for Research Agents Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00065 - MAS-ProVe: Understanding the Process Verification of Multi-Agent Systems Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.03053 - How Correct Is Your Answer? A Semantic Correctness Framework for Open QA Evaluation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01369 - RPCBench: A Benchmark for Proactive Premise Critique in LLM-based Recommendation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00918 - CHARM: Character Hallucination for Multicultural Role Play Benchmark Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01352 - TWIX: a Two-Stage Approach for End-To-End Named Entity Recognition and Relation Extraction Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00832 - Evaluating Second-Order Bias of LLMs Through Epistemic Entitlement Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.17506 - HarnessDev: Can LLMs Create and Evolve Their Own Agent Harness? Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01437 - Beneath the Diff: Diagnosing and Mitigating Algorithmic Mode Collapse in Code-Level Autonomous Research Loops Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00077 - PersianAnonymizer: Evaluating LLM-Labeled Training for Efficient NER-based Anonymization in Persian Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00958 - Disclosure-Gated User Simulation for Companion-Agent Evaluation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00982 - Does task decomposition improve automatic NLG evaluation? Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01139 - VocalAffectBench: Evaluating Vocal Emotion Recognition in AI Audio Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.28932 - StudentSim: Training LLM-based Student Simulators Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01591 - Bridging the English-Arabic Medical Knowledge Gap: Targeted Low-Rank Adaptation via Causal Layer Selection Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.00207 - Vision Is Not Overhead: One-Pass Block Drafting for Lossless Speculative Decoding in Vision-Language Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00355 - What Does an Agentic Software Engineering Benchmark Measure? Profiling Task Demands and Agent Behaviour Beyond What Category Labels Reveal Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01271 - From Detection to Refusal: Safer LLMs via Circuit-Guided Weight Scaling Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00051 - Can Large Language Models Forecast What Researchers Study Next? Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00747 - Whether LLMs Can Navigate Beliefs and Facts Depends on How You Phrase It Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.17809 - Detecting Hidden Behaviors in LLMs via Activation-matched Finetuning Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00351 - Commit-first LLM judging inherits the judge's own errors Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00088 - A Unified Mechanistic Analysis of Knowledge- and Safety-Based Refusals Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00760 - Ready to Speak: Aligning LLMs for TTS-Friendly Text Generation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01246 - WorldBench: Culturally Grounded Benchmark for Multilingual Agents Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01056 - Toppling the Hierarchy in Byte-level Language Modeling Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00463 - Detoxifying Toxic Communication: A Design Science Approach to Responsible AI Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00361 - Verifiable Disaster Storylines and Causal Knowledge Graphs: A Citation-Grounded Pipeline from Heterogeneous Humanitarian Sources Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00858 - Evaluating Style-Personalized Text Generation: Challenges and Directions Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2508.06374 - Explore Before Committing: Hypothesis-Guided Search for Deep Research Agents Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01294 - EDRAC: Benchmarking Arabic Dialect Reading Comprehension Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01113 - FinLifeBench: Exhaustive Life-Event History and Financial-State Reconstruction from Longitudinal Banking Dialogue Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01198 - The Scientific Contribution Graph: Automated Literature-based Technological Roadmapping at Scale Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.15011 - InComeS: Integrating Compression and Selection Mechanisms into LLMs for Efficient Model Editing Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2505.22156 - PlanarBench: Evaluating LLM Spatial Reasoning via Planar Graph Drawing Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.02010 - Hints Help But Do They Teach? Evaluating Skills Transfer in Code Generation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01106 - Compile, Don't Memorize: A Context Compilation Architecture (CCA) for In-Context Learning Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00759 - Oblivion: Self-Adaptive Agentic Memory Control through Decay-Driven Activation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.00131 - Inspicio: Open-Vocabulary, LLM-Based Sense Retrieval for Historical Languages Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00998 - Over-Refusal and Representation Subspaces: A Mechanistic Analysis of Task-Conditioned Refusal in Aligned LLMs Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.27518 - Can Large Language Models Handle Discourse Particles? A Case Study of Colloquial Malay Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.28782 - When Features Become Instances: Inverted Contrastive Learning for Unsupervised Feature Selection Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00782 - Measuring Optimal Transport in Transformer Depth Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00748 - Consistency Without Alignment: Item-Sensitive Language Models Indistinguishable From Random Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00576 - Latent Mechanisms of Language Control in Multilingual Language Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00325 - Retrieval, Scoring, and Decoding Shape Performance and Stability in LLM-based Conversational Recommendation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00086 - A Token is Worth over 1,000 Tokens: Efficient Knowledge Distillation through Low-Rank Clone Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2505.12781 - BOW: Training Language Models to Reason Over Plausible Next Words Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2506.13502 - ClinTraceBench: Source-Verifiable Longitudinal Clinical Reasoning over EHR-Derived Dialogues Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01111 - SafeMath: Safe Solutions for Unsafe Math Word Problems Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.25201 - Multilingual Medical Reasoning for Question Answering with Large Language Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2512.05658 - Control-Data Flow Separation: Stable Prompt Optimization in Multi-Agent LLMs Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00621 - EdiTikZ: Scientific Figure Editing from Revision Trajectories Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01409 - Overfitting Mitigation via Singular Value Decomposition in Minimum Bayes Risk Decoding Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01135 - Life Operators: a self-evolving framework for multiscale life modelling Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00068 - Learning What to Retain: Gated-Memory Routing for Efficient Collaboration in Multi-Agent LLM Systems Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00237 - A Dataset for Modeling Iterative Problem-Solving Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00940 - MELD: Mel-Spectrogram-Based Speech Language Modeling with Discrete Latent Variables Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.29859 - MusTBench: Benchmarking and Advancing Temporal Grounding in Music LLMs Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.29300 - AdaSearch: Balancing Parametric Knowledge and Search in Large Language Models via Reinforcement Learning Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2512.16883 - CordisBench: Can Language Models Reason About Component Lifecycles in Dynamic Agent Harnesses? Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01600 - PersuaRL: Reinforcement Learning-Driven Multi-Expert Selection for Persuasive Dialogue Generation in Insurance Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01188 - From Base Rollouts to RL Reasoning: A Budgeted Search Perspective Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01274 - Medical Causal Hypothesis Verification with Large Language Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00063 - Beyond Token Positions: Safety Alignment Across Denoising Steps in Diffusion Language Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00495 - Beyond Magnitude: Contrastive Routing for Modular Mixture-of-Experts Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01100 - CoLT-Drive: Counterfactual Long-Tail Benchmarking and Knowledge-Preserving Adaptation for Driving Affordance Prediction Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00242 - VIBE-Bench: Evaluating Personalized Large Language Models When Profiles Don't Mean Preferences Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00921 - CaRL-EM: Cost-Aware Reinforcement Learning for Entity Matching with LLMs Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01195 - Chain-of-Thought Faithfulness of Reasoning Models Varies with Where and How Preference Cues Are Delivered Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.29464 - Designing Proactive Thought Partners for Writing Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01588 - From Tool Use to Technological Agency: LoopCAT as a Local-First, Open-Source Tool for Translation Technology Education Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00344 - Enoki: Efficient Multi-Level Hallucination Detection Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00581 - SFAD: Speculative Factuality-Aware Decoding Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00796 - Think Like a Doctor: Conversational Diagnosis through the Exploration of Diagnostic Knowledge Graphs Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.01995 - Probing Factual Knowledge Transfer with Training Data Interventions Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01341 - MiNER: Fine-Tuned Biomedical Natural Language Processing for Malaria Disease Entity Recognition in Clinical Texts Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00073 - GlossoGen: Emergent Language in Complex Multi-Agent LLM Interactions Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01491 - InteractBench: Benchmarking LLMs on Competitive Programming under Unrevealed Information Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.29632 - Latent Recurrent Thoughts: Recurrent Refinement of Proposed Latents for Reasoning with Frozen LLMs Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01117 - GuidedBench: Measuring and Mitigating the Evaluation Discrepancies of In-the-wild LLM Jailbreak Methods Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2502.16903 - From Terminology to Diagrams: Visual-Instruction Generation for Scientific Diagram Understanding Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00948 - Creative Generation via Multi-Agent Debate: Does Debate Suppress Diversity? Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00683 - Polish ModernBERT: The Long and Short of Polish Language Understanding Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01379 - GUI-CC: Benchmarking Contextual Consistency of GUI World Models as Agent Environments Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00048 - Attribute-Based Activation Steering of LLMs for Group-Specific Explanation Generation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.29215 - ALEE: Any-Language Evaluation of Embeddings via English-Centric Minimal Pairs Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2607.00171 - Beyond Tokens: Semantic-Aware Speculative Decoding for Efficient Inference by Probing Internal States Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.03708 - FormalTCS: Benchmarking End-to-End Frontier Formal Theoretical Computer Science Research of Large Language Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.20153 - The Privacy-Hallucination Tradeoff in Differentially Private Language Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00492 - OASIS: Optimizing Attacker Sequences for Hard-Label Black-Box Text Attacks Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.29568 - M4FC: a Multimodal, Multilingual, Multicultural, Multitask Real-World Fact-Checking Dataset Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2510.23508 - Do Multimodal LLMs See Before They Read? Diagnosing Contextual Sycophancy Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00067 - On the Design Fundamentals of Pixel Text Representation Learning Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01147 - Two locked tests of phase-structure features for transition prediction Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00335 - Skill Following: Evaluating Actual Skill Use in Retrieval-Enabled LLM Agents Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00549 - Adaptive Critical Token-Aware Retrieval for Repository-Level Code Generation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01601 - Subword Segmental BabyLMs: Learning to Tokenise for Sample-Efficient Pretraining Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01151 - Argument Quality Assessment with Large Language Models: A Pairwise Bradley-Terry Approach Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.28313 - TEIDAN: A Multilingual Multiparty Dialogue Corpus Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00802 - Automatic Item Generation for Personality Situational Judgment Tests with Large Language Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2412.12144 - LLMBridge: An LLM Pipeline for End-to-end Referential Bridging Resolution in English Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.29048 - Graph Evidence Is Not Enough: Diagnosing Native Decoder Use in Graph-Augmented LLMs Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.30437 - A systematic Approach to constructing a Chance-and-Risk Matrix for Semiconductor Supply Chains Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01563 - LOOMSUM:Weaving Quantitative and Narrative Evidence for Faithful Long Text-Table Summarization Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00241 - The Interlingua Hypothesis: LLMs Translate via a Latent Task-agnostic Feature Space Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00515 - MemeBridge: A Dataset for Benchmarking and Mitigating the Bidirectional Cultural Gap in Meme Interpretation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00491 - Efficient SWE Agent Benchmarking via Trajectory-Aware Evaluation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01603 - Calibration is the Bottleneck: An Action-Class Diagnostic of Multi-Turn Tool-Calling Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00949 - AfriSUD: A Dependency Treebank Collection for Evaluating Models on African Languages Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.12708 - PropUQ-MAS: Propagation-Aware Uncertainty Quantification for LLM Multi-Agent Systems Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.22130 - Padamitra: Grounded Glossary Generation for Classical Sanskrit Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.25038 - Replacing Training with Memory: Listwise Selection for Text-to-SQL Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00834 - PCoMoE: Shifting MoE Inference from Monolithic Expert Selection to Fine-Grained Path Composition Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01024 - Post-hoc Alignment of LLM-judges to Human Judgment Distribution Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01073 - ExpArt-KG: Artwork Image Description Generation through Iterative Exploration of Knowledge Graphs Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00629 - Who Annotates in NLP? A Large-scale Assessment of Human Annotation Reporting between 2018 and 2025 Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.02255 - Exploring Collaboration between a language and a non-language agent Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00474 - Is Human Annotation Necessary? Iterative MBR Distillation for Error Span Detection in Machine Translation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.12983 - LLM-as-a-Demographic: Whom Sociodemographic Prompting Helps, and Whom It Hurts Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00222 - HiDiffTIR: Hierarchical Difficulty-Aware Policy Optimization for Multi-Turn Tool-Integrated Reasoning Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.21863 - Human-Anchored Factuality Evaluation with Strategic Annotation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00494 - SCoNE: Selective Context-aware Neuron Editing for Robust Retrieval-Augmented Generation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00689 - FaST: Feature-aware Sampling and Tuning for Personalized Preference Alignment with Limited Data Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2508.04698 - Energy-Based Transformers as Predictors of Reading Difficulty Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.23382 - KItCAT: Knowledge Injection via Input Corruption for Auto-regressive Training Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00082 - NewsRECON: News Article Retrieval for Image Contextualization Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2601.14121 - One-shot Style Transfer LLM log-probabilities for Authorship Attribution and Verification Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2510.13302 - SpokenUS: A Spoken User Simulator for Task-Oriented Dialogue Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.16783 - SURE-Challenge: Evaluating Speech Evidence Before Speech-LLM Generation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.27783 - The Importance of Being Statistically Earnest: A Critical Re-evaluation of GSM-Symbolic Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.28700 - Emotional Labor Strategy Preferences in LLM Personas Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00310 - TopoAlign: A Framework for Aligning Code to Math via Topological Decomposition Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2510.11944 - SDARE-Bench: Evaluating Large Language Models on Conversational Stigma Detection and Response in Dyadic and Group Dialogue Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01548 - Phrase-Localized Language-Contrastive Guidance: Training-Free Localized Accent Control for Code-Switching Text-to-Speech Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01016 - Slow to See, Slow to Suppress: Understanding the Effects of Modality in Context-Memory Conflicts Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00293 - ChatDev 2.0: A No-Code Multi-Agent Platform for Developing Everything Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00714 - Zero-Shot Respiratory Sound Classification through LLM-Augmented Audio-Text Alignment Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00055 - Instella-MoE Technical Report Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00791 - DigitalCoach: Communication and Grounding Gaps in Human and Agentic Computer Use Coaching Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.31980 - Some Emotions Run Deeper: Layer-wise Probing and Causal Intervention in Large Language Models Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01279 - Deep Thought Alignment: Trajectory-Level Latent Distillation for Video Reasoning Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.16316 - Simple Additions, Substantial Gains: Expanding Scripts, Languages, and Lineage Coverage in URIEL+ Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2510.27183 - Same Semantics, Different Outcome: On the Modality Robustness of Multimodal LLMs under Knowledge Conflict Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00550 - Knowledge Distillation During Mid-Training Favors Reasoning over Factual Recall Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01532 - Arkios: An Open Bilingual English-Nepali Language Model Trained From Scratch, with a Devanagari-Aware Tokenizer Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.30092 - Membership Inference in Fine-tuned Diffusion Language Models via Token-level Memorization Asymmetry Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00873 - XQDT: eXplainable and Quantitative Data-Text Alignment Metric with Feedback Signals Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.29948 - FineVerify: Scaling Test-Time Compute with Fine-Grained Self-Verification for Agentic Search Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.00660 - Separating Syntax from Language: A Mechanistic Account of Translation in Multilingual LLMs Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01356 - SciTrue: Reliable Scientific Claim Validation with Frontier and Open Language Models at the NTCIR SciClaimEval Task Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00654 - trajectory-judge: What Outcome-Only LLM Judges Miss on Agent Trajectories Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00038 - Computational Humor with Multimodal LLMs: Methods, Datasets, Evaluation, and Challenges Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2607.19011 - Beyond Static Summarization: Proactive Memory Extraction for LLM Agents Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2601.04463 - Self-Evolving World Models for LLM Agent Planning Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.30639 - When Modality Gap Reduction Fails: Prediction-Level Hubness in CLIP Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01103 - Auditing Generative Audio Calls for Known-Task Audio-LLM Evaluation Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.27817 - Do as I Say, Not as I Do: Instruction-Induction Conflict in LLMs Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.20382 - Investigating Assistant Bias in LLM User Simulators Using a Role Vector Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00608 - Scope3Trace: Evidence-Based Identification and Extraction of Scope 3 GHG Emissions from Sustainability Reports Source: arxiv-cs-cl Topic: AI Research (+AI) URL: https://arxiv.org/abs/2607.17122 - Delegation Without Trust: An Empirical Gap Analysis of Identity, Authorization, and Runtime Governance in Multi-Agent LLM Systems Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00267 - Towards a Belief-Based World Model for LLM Agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00455 - Defense-as-Skill: Evolving Runtime Guard Skill for Skill-Augmented Agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01487 - AREX: Towards a Recursively Self-Improving Agent for Deep Research Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2607.21461 - Multi-Agent LLM Orchestration Achieves Deterministic, High-Quality Decision Support for Incident Response Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2511.15755 - AnalysisBank: An Expert Analysis Pattern Library for Financial Report Generation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00818 - Successive Capacity Growth: Task-Complexity-Driven Width and Depth Expansion for Vision Transformer Encoders in JEPA World Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.27367 - Drift-Aware LLM Routing with Sparse Contexts and Shared Budgets Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00662 - TRU: Targeted Reverse Update for Efficient Multimodal Recommendation Unlearning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.02183 - Cleaner Speech, Weaker Generalization: Revisiting Pitt-Derived Benchmarks for Alzheimer's Disease Detection Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00276 - Individualized Algorithmic Advice as a Strategic Signal on Competitive Markets Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2511.09454 - Runtime-Independent Persistent Agents: Preserving Identity, Memory, and Code Across Models, Harnesses, and Servers Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00546 - VOIM: Training-Free Open-Vocabulary 3D Instance Mapping for RGB-D and Monocular SLAM Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00775 - On Synthesis of Metric Interval Temporal Logics Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01032 - Feedback-Assisted Trust Propagation over Document Relation Graphs for Retrieval-Augmented Generation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00543 - DramaChain Bench: An End-to-End Benchmark for Short-Drama Generation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00646 - RecalibrateGPT: AI Fatigue Resilient Conversational Interfaces Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00506 - SpecMind: Enabling Spectrum Intelligence via Multi-Agent Hybrid Retrieval-Augmented Generation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00427 - Hypotheses-Guided Self Distillation for Continual Personalization Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00251 - Verifiable abstention makes AI leak diagnosis accountable in urban water distribution networks Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.18836 - EULER: Exploring Underused Links with Evidence-Checked Return for Multi-Agent Mathematical Discovery Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00032 - When Safety Routing Breaks: Understanding Alignment Fragility under Benign Fine-Tuning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01455 - CARE: Contrastive Anchor-based Rubric Evolution for Large Language Model Post-Training Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00892 - mimeo: Compiling Public Expert Corpora into Agent Skills and Testing What Transfers Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00453 - Text-guided flow matching enables sample-efficient crystal structure generation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01076 - AI Should Not Only Be Helpful. It Should Be Contingent. Artificial Intimacy, Sycophancy, and the Future of Social Learning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00211 - ParaStudent: Closing the Sim2Real Gap in User Simulators for AI Tutor Evaluation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2507.12674 - A Composable Evaluation System for Reproducible Omni-Modal Foundation Model Evaluation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01315 - Same Request, Different Boundary: Evaluating Cybersecurity Assistance across Conversational Contexts Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00578 - Beyond the Clock: Measuring the Value of Adaptive Revision Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00874 - The Lifecycle of LLM-as-a-Judge for Large-Scale Recommendation Explanations Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.18300 - Neural-Primitive: An Efficient End-to-end Local Planner with Primitive-based Imitation Learning for Autonomous Flight Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.20948 - User Representation via Cross Multi-source Behavior Pre-training for Mobile Games Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01057 - SeerGuard: A Safety Framework for Mobile GUI Agents via World Model Prediction Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2607.15550 - The Answer Is Not the Argument Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00264 - Towards Generalizable Visually Grounded Exploration of Household Devices Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00845 - QILP-0: Constructing Observational Declarative Twins of Quantum Circuits Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01049 - X-SG$^2$S: Safe and Generalizable Gaussian Splatting with X-dimensional Watermarks Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2502.10475 - AI Morbidity and Mortality: A Framework for Clinical AI Failure Review Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00076 - Diffusion Large Language Models for Visual Speech Recognition Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.28456 - Neuro-Symbolic Geometric Abstraction (NeuSOGA): From Observations to Symbolic Mathematical Representations Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01408 - H3-World: Turning Language Understanding into World Control Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01560 - ARISE-RL: Agentic Rubric-Grounded Iterative Self-Evolution with Reinforcement Learning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01058 - Asymmetries in Spontaneous and Instructed Deception Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00180 - MicroEvo: Knowledge-Guided LLM Sampling for Efficient Microarchitecture Design Space Exploration Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.06183 - Dual Process Motion Planning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01260 - CoBRA: Learning Tool-Use Boundaries via Counterfactual Margins Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00967 - The Artificial Experimentalist: Discovery and Control of Self-Organizing Phenomena with Autotelic Reinforcement Learning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.26116 - Probabilistic Model Checking of Autoregressive Neural Sequence Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00838 - Wave Function Backpropagation with Explicit Temporal-Interval Dynamics Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00503 - TimeSteer: Inference-Time Speech Scheduling in Joint Audio-Visual Diffusion Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01277 - Space Generative AI with Solar Energy Harvesting Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01062 - Invalidation Contracts for Cross-Episode Agent Memory Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00243 - Making Prospective Memory SLM-Shaped: Typed Intention Stores for Small-Model Agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01272 - HiveTraceGuard-Pro: A Compact Generative Guardrail for Prompt Injection, Jailbreaks, and Adversarial Obfuscation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01046 - PopPert: Population-level Joint-Distribution Modeling for Single-Cell Perturbation Prediction Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01357 - Conversation Coach: A Voice-enabled AI System that Helps Practice Difficult Workplace Conversations Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00441 - Operational Regimes in Non-Convex Optimization: A Multiplier-Based Taxonomy Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00471 - A Mathematical Framework for Legacy, Governance, and Decision Integrity in Enterprise AI Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00572 - H2Table: Hierarchical Hypergraph-Enhanced Large Language Models for Complex Table Reasoning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01216 - Reconstruct! Don't Encode: Self-Supervised Representation Reconstruction Loss for High-Intelligibility and Low-Latency Streaming Neural Audio Codec Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.05887 - pro-team at LLMs4OL 2026 Tasks Flagship and Reuse: Retrieval-Augmented Generation and Vocabulary-Constrained Filtering for Ontology Learning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.27101 - Training-Free Refinement of Flow Matching with Divergence-based Sampling Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.04646 - Agentic programs: an emerging form of scientific software in computational materials science Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00795 - SoK: When Safe Agents Fail Together: The Security of Multi Agent LLM Systems Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00595 - Reasoning-supported Robustness Validation of Automotive E/E Components Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.16421 - DNC-IMM: Early Lane-Change Intention Recognition via Neural Calibration Based on Driving Context Information Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01120 - DiagEvo: Diagnosis-Guided Self-Evolution via Hierarchical Error Memory Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00768 - In-Context Neurofeedback: Can LLMs Control Their Internal Representations through Privileged Access? Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00904 - LEAP: Likelihood Elicitation and Aggregation for LLM-based Probabilistic Forecasting Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01337 - Jailbreaking Text-to-Image Models Through Cracks: Navigating Heterogeneous Safety Filters via Multi-Agent Debate Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01168 - Event-Aligned Analysis of Multi-Rater Pain Assessments Using Continuous Wearable Physiology Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.23705 - The zbMATH Open Knowledge Graph: Tracing Centuries of Mathematical Research Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00969 - Reinforcement Learning Enhanced LLM Agents for Complex Vehicle Routing Problems Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00859 - Incremental Risk Assessment of Progressive Elder Financial Scams via Instruction-Tuned Small Language Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00005 - OpenAgentFlow: Enabling System-Wide Safety Boundaries for Heterogeneous AI Agent Fleets Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00015 - Cost-efficient Active Learning for Referring Image Segmentation and Grounding Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.30621 - Socrates went Nuclear: Comparing Interaction Strategies for AI systems in a Learning Context using Brain Sensing Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00584 - Athena: Vulnerability-Affected Library Identification via Knowledge Graph Completion Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01187 - V-Co: A Closer Look at Visual Representation Alignment via Co-Denoising Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.16792 - A Machine Learning-Driven Solution for Denoising Inertial Confinement Fusion Images Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2511.16717 - The Safeguard Worked. Is the LLM System Safer? Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00519 - Restrict, Don't Retrain: Inference-Time VLM Guidance for Zero-Shot Aerial Segmentation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00628 - Uncovering the Computational Ingredients of Human-Like Representations in LLMs Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2510.01030 - Autonomous discovery of new structure-plausibility laws for explainable and rapid crystal diagnosis and screening Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01209 - Guided Prompt Evolution for Vision-Language Models Adaptation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.09493 - SCAFFOLD: A Large-Scale Structured Dataset of Computer Science Research Figures with Diagram QA and Chain-of-Thought Reasoning Traces Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00018 - Prompt-Robust Language Models: Which Training Strategies Work? Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01217 - FoldingAgent: Inferring Parametric Origami Procedures from Demonstration Videos Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00377 - RAPIDMap: Rapid Multi-Agent Pipeline for Interpretable Disaster Mapping from Satellite and Street-view Imagery Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00046 - KGFR: A Foundation Retriever for Generalized Knowledge Graph Question Answering Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2511.04093 - When Guardrails Look Effective: Construct Validity Failures in LLM Agent Commerce Evaluation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01519 - Figures as Programs: Recursive Generation of Editable Scientific Figures Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01006 - REVISE: Validity-Guided Recovery for Online Revisions in Agent Workflows Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00643 - IWP: Token Pruning as Implicit Weight Pruning in Large Vision Language Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.00757 - StainPresetNet: Stain Preset Network for Fast Multi-to-Multi Stain Normalization Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01146 - SkillRet: A Large-Scale Benchmark for Skill Retrieval in LLM Agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.05726 - AgentFactory: Towards Automated Agentic System Design and Optimization Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01045 - AutoXRD: Autonomous LLM Agents and Comprehensive Evaluation for Powder Diffraction Analysis Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00070 - TUTTI: Toward generalizable audio-to-score transcription via fully synthesized data Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00640 - StudyBench: Can Self-Evolution Squeeze Textbooks for Olympiad Capability? Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00787 - The Assistant's Ideal Self Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00304 - Long-Horizon State Tracking in LLMs: Executing MD5 through a Deep Sequence of Dependent Tool Calls Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00012 - LUNA: Learning Universal 3D Human Animation Beyond Skinning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.31981 - HiLRP: Toward One Trustworthy Explanation for Vision Transformer: Conservation-Valid Attribution via Attention Primitives Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01282 - FLaG: Frequency-Domain Latent-attention Gated Pooling for Token Aggregation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00831 - EvoSCM: Scientific Belief Revision Through Causal Model Evolution and Experimentation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01526 - ContextPipe: Database-Inspired Context Assembly for Long-Horizon Agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00749 - UniACE: A Unified Framework for Evaluating LLM Agentic Capabilities Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.27898 - A Formal Analysis of Agent Payment Protocols Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00060 - Mechanism Design for Alignment and Control Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01595 - Electronic Navigational Chart Change Classification Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.20218 - EGT-KG: Evidence-Grounded Typed KG Retrieval for Practical Scientific QA with Small Language Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00479 - Dependency-Aware Chain-of-Thought Compression for Financial Reasoning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00413 - One Policy, Any Budget: Internalizing Budget-Aware Search via Reinforcement Learning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00813 - Human-AI Co-Interpretation for Responsible AI: A Hermeneutic Perspective Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00334 - Agentic Large Language Models for Training-Free Neuro-Radiological Image Analysis Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.16729 - Can LLMs Design Video Coding Tools? A Case Study on Planar Mode Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01535 - One Prompt Is Enough: Watermark Laundering Through Foundation Image Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01249 - SlideBank: A Persistent Hierarchical Evidence Bank for Consistent Whole-Slide Reasoning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00342 - VectorGym: A Multi-Task Benchmark for SVG Code Generation, Sketching and Editing Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.29852 - Does Reasoning Mitigate Backdoor Attacks? A Neuro-Symbolic Perspective Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00464 - ReDeck: Step-Level Render-Grounded Refinement for Document-to-Slide Generation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00194 - Flow Reasoning Models: Turning Flows Into Efficient Recurrent Reasoners Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.29150 - UrbanDS: A Graph-Guided LLM Multi-Agent System for Data-Intensive Urban Tasks Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2607.26724 - DiffHDR: Re-Exposing LDR Videos with Video Diffusion Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2604.06161 - Make Some Noise: Unsupervised Remote Sensing Change Detection Using Latent Space Perturbations Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2602.19881 - Rock, Paper, Scissors, ... Dynamite - A Model of Disruption from New Technologies Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00207 - Discrete-Time MDP Modeling for Multi-Item Capacitated Lot Sizing with Stochastic Demand Timing Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00004 - Causal Evidentiary Governance for High-Risk Machine Learning Systems Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01040 - Does Runtime Topology Context Improve LLM-Generated Kubernetes Security Patches? Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2607.25995 - When the Algorithm Becomes the Brand Crisis: A Sociotechnical Theory of Distributed Responsibility and Accountable Transparency Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00510 - Distributed Implicit Harm: A Compositional Safety Blind Spot in MLLM-Based Video Moderation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00206 - Automated Tree Knowledge Graph Construction using Ontology Expansion and Retrieval from Vietnamese History Textbooks Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00763 - IMPACT: Attention Is the Interaction Map for Scalable Interaction-Aware World Model Training Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00161 - Relational-Core Graph Analytics Querying graphs at SQL scale, and why the node/edge model is a performance tax, not a truer picture of connected data Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01525 - ViPlan: A Benchmark for Visual Planning with Symbolic Predicates and Vision-Language Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2505.13180 - Autoresearch for Marketplace Catalogs: From Legacy Forms to AI-Native Matching Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00274 - On the Human and Computer Alignment of Attribute-Based Music Matches Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00987 - Few-Shot Out of Domain Intent Detection with Covariance Corrected Mahalanobis Distance Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00961 - Escaping Redundant Reasoning: Structure-Aware Search for Inference-Time LLMs Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00738 - Measuring the Behavioral Fidelity of Long-Horizon Human Activity Simulations Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01257 - A Network Science Perspective on Evaluating Deep Graph Generative Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01015 - Generativism: Toward a Learning Theory for the Age of Generative Artificial Intelligence Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.12441 - Towards reliable multimodal disaster severity assessment through preference optimization and explainable vision-language reasoning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00879 - Autoregressive Mosaics: Probing 2D Spatial Reasoning in Text-Only Language Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.30751 - EmbodiedSkills: A Unified Framework for Orchestrating, Training, and Deploying VLA Agents Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01281 - CoVer: Conflict-Aware Claim Verification Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00508 - Automated Event Log Generation from Unstructured Text Using Finetuned LLMs Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01320 - Causal Probing for Internal Visual Representations in Multimodal Large Language Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2605.05593 - MADS: A Multiview Acoustic Descriptor Set Beyond Standard Spectral Summaries Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00792 - A Hybrid Insider Threat Detection Framework Combining Multi-Agent Simulation, Layered SIEM Correlation, and Theory-of-Mind Reasoning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2601.04243 - Solaris: Towards Interfaces That Are Generated, Not Coded Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00776 - HyperWorld: Hypergraph-Structured State Serialization Improves Learned Textual World Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00002 - TempCloze: Can Video-LLMs Identify the Missing Middle? Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01515 - Accelerating Unified Multimodal Models with Core-Expansion Routing and Unified Computation Scheduling Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.29291 - S^3martCirc: Self-supervised Smart Circuit Discovery Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00755 - WiseSpec: Requirements-Driven Agents for Code Generation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00568 - LifeAgentBench: Benchmarking LLMs for Long-Horizon, Cross-Dimensional Lifestyle Health Reasoning Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2601.13880 - Investigating Hyperparameter Optimization and Transferability for ES-HyperNEAT: A TPE Approach Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00449 - SARTM: Segment Any RGB Thermal Model with Language aided Distillation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2505.01950 - Semi-Supervised Virtual Staining via Morphology Preservation and Histopathological Realism Constraints Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00984 - Recursive Criticality of AI Self-Improvement Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00137 - When Oracle Conditioning Misleads Deployment: Conditioning-Availability Bias in Echocardiographic Segmentation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.03342 - GeoGR^2:Zero-Shot Geospatial Inference via Geostatistically-Guided Iterative Refinement with LLMs Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2508.04080 - Intelligent Edge Computing Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00181 - Visual Attention Faithfulness in Vision-Language Models is Heterogeneous Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00830 - The Irreversibility Budget: Fleet-Level Risk Accounting and Admission Control for Agent Operating Systems Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00275 - Don't Let the Model Write the YAML: Deterministic, Minimal-Diff GitOps Remediation from LLM-Proposed Field Changes Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00227 - EDGE: Error Dependency Graph-Guided Multi-Error Attribution in Multi-Agent LLM Systems Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01360 - A Human-AI Theorem Connecting Spontaneous and Field-Induced Mechanisms of Collective Behavior in One Dimension Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00322 - BS: Take the Hint - Interactive Multitracer PET/CT Lesion Segmentation with a Scribble-Conditioned ResEnc U-Net Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01554 - ConvDeck: Conversational Paper-to-Slide Generation via Stage-Specific User Feedback Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00226 - Heard but Not Heeded: Paralinguistic Information Encoding and Loss in Audio-Language Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00727 - ISO-RAG: Isoperimetric Noise Control for Retrieval-Augmented Generation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00513 - Beyond the Image Plane: World-Grounded Queries for Multi-Object Tracking Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00924 - Revealing Multi-View Hallucination in Large Vision-Language Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2603.23934 - Parsing the Stream: A Live Trace Model for Long-Horizon Agents and Their Observers Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01466 - ADGNet: Asymmetric Dual-text Guided Network for Infrared Small Target Detection Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00853 - Self-EmoQ: Plutchik-Guided Value-based Planning to Drive Streaming Emotional TTS Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.09837 - SymFold: Synergizing Evolutionary and Structural Priors for Accurate Protein Inverse Folding Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01353 - Data-Driven Persona-Conditioned Agents for A/B Test Simulation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01038 - Residual Sparsification via Output Importance for Compressing Mixture-of-Experts LLMs Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00575 - Revisiting Face Recognition for Monozygotic Twins: The Celeb Twins Test Set Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01141 - Evaluating Multimodal LLMs as Generalist Vision-Language-Action Agents for Drone Control: Commanding, Approaching, Tracking and Searching Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01404 - Harness-of-Harness: Multi-Day Autonomous Software Development with Continual Improvement Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01481 - MutMem-V2: Cryptographically Authorized Mutation in Persistent Agent Memory Portable Verification and Reproducible Evidence Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01235 - Differentially Private Paired Table-Image Multimodal Synthesis Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00708 - CacheBridge: Efficient Cross-Model KV Cache Transfer Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00891 - Are We There Yet? Assessing Computer-Use Agents for Blind Users' Accessible Interaction with Desktop Applications Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00524 - Analog-DB: An Agent-First Analog Integrated Circuit Database, From Blocks to Systems Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01286 - Validity-Aware Jailbreak Evaluation for Large Language Models Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00498 - Triple-Bottom-Line Sustainability of Language Models for Edge AI: A Comparison Between SLMs and Quantized LLMs Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00665 - Scaling Large Reasoning Models beyond Human Supervision: A Path toward Superintelligence Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.31075 - Position Matters: Feature Inversion Attacks in ViT Split Inference with Token Reduction and Shuffling Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01232 - Deploying and Evaluating a Smart-Agriculture Agentic Engine for Full-Season Soybean Farm Operations Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00106 - RestoreBench: Can AI Agents Restore Power Flow Convergence? Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00384 - Towards a Reliable and Practical Eval Pipeline Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00805 - Authority Bias in Conversational Search Engines for Academic Paper Recommendation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00248 - GeM-NR: Geometry-Aware Multi-View Editing for Nonrigid Scene Changes Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2606.05142 - Taming Modality Entanglement in Continual Audio-Visual Segmentation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2510.17234 - RACE: Scalable Statistical Estimation of Functional Consistency in LLM Neurons Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2608.24758 - A Closed-Loop Evaluation of Capability Loss and Recovery in Compressed Driving Policies Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00718 - Scalable Rao-Blackwellized Online Planning for High-Dimensional POMDPs Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01351 - Benchmarking Vision-Language Models for Automated Pathology Diagnosis and Report Generation Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.00866 - Who Judges the Judges? A Chinese Safety QA Benchmark for Evaluating LLM Responses and Safety Judges Source: arxiv-cs-ai Topic: AI Research (+AI) URL: https://arxiv.org/abs/2609.01210 ## Community - llm-gemini 0.34 Source: simon-willison Topic: Python (+AI, Django) URL: https://simonwillison.net/2026/Sep/2/llm-gemini/ - Claude's new system prompt really doesn't want to reproduce song lyrics Source: simon-willison Topic: Python (+AI, Django) URL: https://simonwillison.net/2026/Sep/2/claudes-new-system-prompt/ - Quoting Rick Brewster Source: simon-willison Topic: Python (+AI, Django) URL: https://simonwillison.net/2026/Sep/2/rick-brewster/ - New to Rust, converting a project from Python Source: rust-users-forum Topic: Rust (+AI) URL: https://users.rust-lang.org/t/new-to-rust-converting-a-project-from-python/142184 - METR Report on OpenAI / Hugging Face Hacking Incident Source: hacker-news Topic: General (+AI) URL: https://metr.org/blog/2026-08-26-openai-hugging-face-incident-investigation/#core-takeaways-about-this-incident - Justice Dept. Sides with OpenAI in New York Times Copyright Suit Source: hacker-news Topic: General (+AI) URL: https://www.nytimes.com/2026/09/02/technology/justice-department-openai-copyright-suit.html - Launch HN: RonanRX (YC S26) – Personalized Peptides and GLP-1s Source: hacker-news Topic: General (+AI) URL: https://ronanrx.com/ - Show HN: Open source ML programming language playground Source: hacker-news Topic: General (+AI) URL: https://sw-ml-study.github.io/sw-mlpl/ - Gemini 3.8 Flash and 3.8 Flash Cyber Source: hacker-news Topic: General (+AI) URL: https://blog.google/innovation-and-ai/models-and-research/gemini-models/3-8-flash-and-3-8-flash-cyber/ - WebLLM: high-performance in-browser LLM inference engine Source: hacker-news Topic: General (+AI) URL: https://github.com/mlc-ai/web-llm - Six curl CVEs after OpenAI and Anthropic came back with zero Source: hacker-news Topic: General (+AI) URL: https://aisle.com/blog/aisle-discovered-six-curl-cves-after-openai-and-anthropic-found-zero - LLMs: Intelligence vs. Cost Source: hacker-news Topic: General (+AI) URL: https://openteams.com/intelligence-vs-cost/ - Bye Bye Perspective API: Lessons for Measurement Infrastructure in NLP, CSS and LLM Evaluation Source: lobsters Topic: General (+AI) URL: https://arxiv.org/abs/2604.25580 ## Media - End-to-end agentic development with Google Gemini, Antigravity, and GitLab Source: youtube-googlecloud Topic: AWS (+AI, Cloud, General) URL: https://www.youtube.com/watch?v=IXc9d_EN4ts - He built a voice controlled drone to avoid using a remote! 🤯✈️ Source: youtube-googlecloud Topic: AWS (+AI, Cloud, General) URL: https://www.youtube.com/shorts/afUQpY5oYcw - Don't wait to graduate: A student’s AI ancestry storyteller Source: youtube-googlecloud Topic: AWS (+AI, Cloud, General) URL: https://www.youtube.com/shorts/2TQuelvaWMo - Training Agents 4: From reward functions to environments. Source: youtube-huggingface Topic: Hugging Face (+AI) URL: https://www.youtube.com/watch?v=nJV3yUuz6DU - The most interesting hack in history just got weirder... Source: youtube-fireship Topic: Web (+AI, General, JavaScript) URL: https://www.youtube.com/watch?v=0Rp9KJCEIvg - Demystifying AI terms: loop engineering, squads, and harness | S02E02 | The GitHub Podcast Source: youtube-github Topic: Open Source (+AI, General) URL: https://www.youtube.com/watch?v=7oqYIRbB6Rc