AI Research

😺 Claude Fable 5.1 can do the work. The hard part is managing it.

Avoiding Entity Key Drift in a Data Lake: Step 2, When Fuzzy Matching Stops Working

I built a matcher meant to finish the cleanup that normalization left behind. Testing it against real data showed that no version of it could be made safe. What follows is the architecture that was left once the matcher…

A RAG That Says “Not in This Document” Has to Show Four Kinds of Evidence

Enterprise Document Intelligence [Vol.1 #B3] - A confident wrong answer is a bug. A bare “no answer” with no justification is almost as bad. Each of the four bricks has one piece of evidence to show The post A RAG That…

Best AI Presentation Makers for Business in 2026

Compare the best AI presentation makers for work, including Gamma, Canva, Plus AI, and Adobe Express, with pricing, use cases, and key tradeoffs.

Graph Neural Networks: GCN, MPNN, and GAT, Explained Simply

A visual guide to how graph neural networks work under the hood The post Graph Neural Networks: GCN, MPNN, and GAT, Explained Simply appeared first on Towards Data Science.

A Practical Introduction to PySpark Window Functions

Why the standard groupBy function isn’t enough The post A Practical Introduction to PySpark Window Functions appeared first on Towards Data Science.

😺Anthropic launched Fable 5.1: and now, the agents cost less

Test-Time Scaling in Reasoning LLMs: Inference Regimes, Evaluation, and Reproducibility

arXiv:2608.04001v2 Announce Type: replace Abstract: Large language models can solve harder reasoning problems with more inference-time compute. The term "test-time scaling," however, covers several inference algorithms:…

A Classifier That Teaches Itself: Self-Improving, Frozen-gate Training (SIFT) for Dynamic Document Classification

arXiv:2607.18358v2 Announce Type: replace-cross Abstract: Document classification is a solved problem in the laboratory and an unsolved one in the enterprise. The blocker is rarely model architecture; it is the labeling…

Convergence issues in Relational Concept Analysis based on AOC-posets

arXiv:2609.00054v1 Announce Type: new Abstract: Formal Concept Analysis (FCA) is an approach for conceptual classification building and rule discovery from a binary table describing a set of objects by a set of…

Soft-Argmax for the Projective Plane via the Veronese Embedding

arXiv:2609.00521v1 Announce Type: cross Abstract: From horizon detection to fibre structures in X-ray imaging, many vision tasks recover lines via peak detection in Hough space $H=S^1\times\mathbb{R}$, the domain of…

Online Self-Weighted Fine-Tuning

arXiv:2609.00734v1 Announce Type: new Abstract: Standard supervised fine-tuning (SFT) assigns the same explicit loss weight to every expert demonstration, regardless of the model's changing competence over training…

EvoFlint: An Evolutionary Atlas of Multi-Turn LLM Vulnerabilities

arXiv:2609.00487v1 Announce Type: cross Abstract: Frontier language models that refuse harmful single-turn prompts often comply when the same intent is reached gradually over many turns, making multi-turn attacks one of…

Why Multi-Layer Message Passing Works: Completeness Theory for Graph Neural Network Interatomic Potentials

arXiv:2609.00528v1 Announce Type: new Abstract: We prove that the Hypergraph Neural Network, an invariant architecture with 3-body message passing, is a universal approximator for potential energy surfaces. Our main…

MineDraft: A Framework for Batch Parallel Speculative Decoding

arXiv:2603.18016v3 Announce Type: replace-cross Abstract: Speculative decoding (SD) accelerates large language model inference by using a smaller draft model to propose draft tokens that are subsequently verified by a…

Multi-View Causal Discovery without Non-Gaussianity: Identifiability and Algorithms

arXiv:2502.20115v4 Announce Type: replace Abstract: Causal discovery is a difficult problem that typically relies on strong assumptions on the data-generating model, such as non-Gaussianity. In practice, many modern…

Right Frame, Wrong Rule: Cultural Cues Expose the Financial Knowledge Gap They Were Meant to Close

arXiv:2609.00999v1 Announce Type: cross Abstract: When a question has valid answers under different normative frameworks, a language model must decide which framework to use and whether it can answer correctly within…

Process-Aware AI for Rainfall-Runoff Modeling: A Mass-Conserving Neural Framework with Hydrological Process Constraints

arXiv:2603.25093v2 Announce Type: replace Abstract: Machine learning models can achieve high predictive accuracy in hydrological applications but often lack physical interpretability. The Mass-Conserving Perceptron…

Scaled Idempotence in Transformer Attention: Paired OV Geometry and Shared-Value Algebras

arXiv:2609.01129v1 Announce Type: new Abstract: We identify a recurrent algebraic regularity in Transformer attention: a sparse subset of effective OV operators $T=OV^\top$ nearly closes under composition,…

Semantic-Guided Multimodal Preprocessing for Vision Transformer-Based Clear Cell Renal Cell Carcinoma Grading

arXiv:2609.01426v1 Announce Type: cross Abstract: Clear cell renal cell carcinoma (CCRCC) grading is essential for treatment planning, yet existing approaches either analyze patch-level images directly or focus solely…