i-2-6

Documents about i-2-6

• Renormalising Generative Models for Active Inference: Foundations, Derivations, and Verification#793840arXiv (All)
• Compression-Aware Abstention: Teaching LLMs to Refuse When KV-Compression Masks Remove Answer Evidence#797760arXiv (OAI Expanded)
• Compression-Aware Abstention: Teaching LLMs to Refuse When KV-Compression Masks Remove Answer Evidence#799183arXiv (All)
• RecEvolve: A Knowledge-Driven Autonomous Agent System for Recommender Systems#822062arXiv (All)
• Sparse Readout Prism: Explaining Logit-Lens Scores in Features Instead of Tokens#822353arXiv (All)
• NOTAI.AI: Explainable Detection of Machine-Generated Text via Curvature and Feature Attribution#927247arXiv (All)
• Emergent Charging Coordination in Electric Delivery Fleets#984594arXiv (All)
• Aligning Language Models with Observational Data: Opportunities and Risks from a Causal Perspective#999659arXiv (All)
• Compressive sensing inspired self-supervised single-pixel imaging#1012172arXiv (All)
• Regularized Coupling Maps on Matrix Polytopes: From Assignment to Graph Matching and Optimal Transport#790086arXiv (OAI Expanded)
• Regularized Coupling Maps on Matrix Polytopes: From Assignment to Graph Matching and Optimal Transport#794105arXiv (All)
• OR-VSKC: Resolving Visual-Semantic Knowledge Conflicts in Operating Rooms with Synthetic Data-Guided Alignment#183270arXiv (OAI)
• Non-Stationarity Breaks Permutation Surrogates in Multi-Agent Reinforcement Learning: Diagnosis and Remedies#986848arXiv (All)
• Signal or Noise? Auditing Rotation-Induced Saliency Drift in Medical and Aerial Imaging#822616arXiv (All)
• Dr. Jekyll and Mr. Hyde: Two Faces of LLMs#182992arXiv
• Dr. Jekyll and Mr. Hyde: Two Faces of LLMs#183042arXiv Biology
• Dr. Jekyll and Mr. Hyde: Two Faces of LLMs#183092arXiv (OAI Expanded)
• Dr. Jekyll and Mr. Hyde: Two Faces of LLMs#183142arXiv (OAI)
• Quipu: A Governed Bitemporal Knowledge Graph Store#623091arXiv (OAI Expanded)
• Quipu: A Governed Bitemporal Knowledge Graph Store#625364arXiv (All)
• Fast 3D Foundation Model Initialized Gaussian Splatting#608923arXiv (OAI Expanded)
• Fast 3D Foundation Model Initialized Gaussian Splatting#609985arXiv (All)
• Efficient Context-Limited Telescope Bibliography Classification for the WASP-2025 Shared Task Using SciBERT#822084arXiv (All)
• LaGSplat: Inferring Physics-Governed Interactive Simulation from Monocular Video Using Latent Lagrangian Gaussian Splatting#619045arXiv (OAI Expanded)
• LaGSplat: Inferring Physics-Governed Interactive Simulation from Monocular Video Using Latent Lagrangian Gaussian Splatting#620724arXiv (All)
• Reinforcement Learning for Heterogeneous Sensor Selection in Maritime Surveillance#821961arXiv (All)
• Short-Window Sliding Learning for Real-Time Violence Detection via LLM-based Auto-Labeling#920845arXiv (OAI Expanded)
• Short-Window Sliding Learning for Real-Time Violence Detection via LLM-based Auto-Labeling#925079arXiv (All)
• Technical Manual for a Toolkit for Measuring Contextual Individuation in Transformer Language Models#972265arXiv (All)
• RAFM-SER++: A Lightweight Multimodal Emotion Recognition Framework for Real-Time Behavioral Monitoring in Surveillance Systems#984335arXiv (All)
• SAPE: Sandwich Adapters for Parameter Efficiency in Large Language Model Fine-Tuning#614363arXiv (OAI Expanded)
• SAPE: Sandwich Adapters for Parameter Efficiency in Large Language Model Fine-Tuning#615330arXiv (All)
• Domain-Grounded Tool Orchestration for LLM-Guided Scientific Analysis#808759arXiv (OAI Expanded)
• The Axiom of Consent: Authorization, Friction, and Multi-Agent Coordination#809298arXiv (OAI Expanded)
• Domain-Grounded Tool Orchestration for LLM-Guided Scientific Analysis#818751arXiv (All)
• The Axiom of Consent: Authorization, Friction, and Multi-Agent Coordination#819290arXiv (All)
• Safety for Whom? Boundary-Aware Self-Distillation for Controlled LLM Safety Refusal#927768arXiv (All)
• Beyond Linear Context: Graph-Guided Evidence Navigation for Long-Novel Reasoning with a Local 9B Language Model#1037622arXiv (All)
• Tricarbonylrhenium(I) halide complexes of 2-(4R,6R)-4,6-dimethyl-1,3-dioxan-2-ylpyridine (L1) and 2,6-bis(4R,6R)-4,6-dimethyl-1,3-dioxan-2-ylpyridine (L2): structure and solution stereodynamics#167993Birkbeck Institutional Research Online
• Tricarbonylrhenium(I) halide complexes of 2-(4R,6R)-4,6-dimethyl-1,3-dioxan-2-ylpyridine (L1) and 2,6-bis(4R,6R)-4,6-dimethyl-1,3-dioxan-2-ylpyridine (L2): structure and solution stereodynamics#168540Birkbeck Institutional Research Online
• InstantMimic: A High Performance System for Learning Physics-based Skills in Seconds#996640arXiv (All)
• ENEAS: Embedding-guided Neural Ensemble for Adaptive Segmentation#926643arXiv (All)
• ADELE - Adaptive Delaunay Grids for High-Fidelity Mesh-Native Reconstruction#974818arXiv (All)
• Empirical Evaluation of Data Poisoning Attacks in Supervised Learning#998525arXiv (All)
• Node-wise Feature Encoding for Neural Performance Prediction#779710arXiv (OAI Expanded)
• Node-wise Feature Encoding for Neural Performance Prediction#781085arXiv (All)
• CAGLIARI 129#930739eCrystals
• Transfer Learning for Matrix Completion#1037729arXiv (All)
• Interpretable Cross-Lingual Alignment in Small Language Models: Probing Cultural and Pragmatic Reasoning in Japanese-English Bilingual LLMs#609313arXiv (OAI Expanded)
• When Single-Dataset Conclusions Fail: A 45-Task Study of Threshold Tuning and Resampling for Imbalanced Classification#618881arXiv (OAI Expanded)