AI News

KGOT: Unified Knowledge Graph and Optimal Transport Pseudo-Labeling for Molecule-Protein Interaction Prediction (arxiv.org)

Hallucinations of Large Language Models in Cross-Language Code Translation: Phenomena and Mitigation (arxiv.org)

VISH-ASR: Viterbi-Guided Selection of Hotwords for Speech-LLM Contextual ASR (arxiv.org)

An LLM-Based Agent for Legal Compliance Assessment: Architecture, Memory, and Evaluation in a Canadian and Quebec Context (arxiv.org)

Fundamental Impossibility Results in AI Safety: Trading Off between Desiderata (arxiv.org)

On the Limits of Scale: Inverse Scaling in Vision-Language Models (arxiv.org)

Exposing Vulnerability Monoculture Through Systemic Adversarial Robustness (arxiv.org)

A Likely Geometry of Generative Models (arxiv.org)

Refuses When Asked, Complies When Acting: A Runtime Safety Check for Robot Control Loops (arxiv.org)

Steering Vision-Language Models for Visual-Contextual Privacy in Location Disclosure (arxiv.org)

Near-Duplicate Image Selection Transfers Preferences in Vision-Language Models (arxiv.org)

The Shape of the DPO Hyperparameter Landscape: What to Measure, How It Moves with Scale, and When Not to Transfer It (arxiv.org)

MLIPS: Multi-Layer Indexer Pool Sharing (arxiv.org)

Fixed-Size Active Statistical Inference (arxiv.org)

Sample-efficient learning with Dense Associative Memories (arxiv.org)

Shaped SAEs: Expanding the Toolkit for Mechanistic Interpretability (arxiv.org)

Trail-Weighted Influence Scores for Approximate Outcome Prediction in Bayesian Networks (arxiv.org)

Selecting the Best Few with Human Evaluators: Adaptively Combining Ratings and Pairwise Comparisons (arxiv.org)

VLA-Retain: Benchmarking Cross-Dimension Effects of Vision Tuning in VLAs (arxiv.org)

CovNav: Conformal Prediction-Guided Adaptive Inference for Object Goal Navigation (arxiv.org)

A Unified Attack Framework against Statistical Generative Watermarks for Large Language Models (arxiv.org)

Corpus Quality Improves, Retrieval Utility Does Not: An Ablation of Preprocessing and Refinement in a Speech-to-QA Pipeline (arxiv.org)

A Head Camera Is (Almost) All You Need: An Electronics-Free UMI Glove Labelled and Observed by One Moving Camera (arxiv.org)

Predictive Closure: A Set-Level Characterization of JEPA's Feature Selection (arxiv.org)

LiBRA: Logit Fingerprints for Byzantine-Robust Client Selection in Federated Learning (arxiv.org)

Automated Brick Schema Knowledge Graph Completion Across Buildings (arxiv.org)

VPFlow: Value-Predictive Planning with Imperfect World Models for Offline Goal Reaching (arxiv.org)

Local Allow, Global Deny: Why Per-Action Authorization Does Not Compose in LLM Agents (arxiv.org)

ABench: Adversarily Benchmarking Large Language Models across Taxonomized Values (arxiv.org)

WORTH Benchmark: Evidence Discovery under Not-Fully-Cooperative Information -- A Two-Episode Pilot (arxiv.org)

Benefit and Cost of Random-Token Replacement under Repeated Data: A Known-Entropy Decomposition and Evidence from Ten Real Corpora (arxiv.org)

AudiPsy: Evidence-Grounded Multimodal Reasoning over Spoken Counseling Conversations for Depression Assessment (arxiv.org)

Learned Readout Tokens for Extracting Large Language Model Confidence (arxiv.org)

AnomalyJev: Towards Evaluating Image Anomaly Explanations via Claim-Level Verification (arxiv.org)

Generalizing Multimodal Foundational Models in Context-Driven Class Incremental Learning (arxiv.org)

When Do Humans and Machines Truly Complement Each Other? Formal Conditions and Empirical Evidence for Human-AI Synergy (arxiv.org)

Hybrid Momentum Methods for Optimizing Deep Learning Models (arxiv.org)

Neural Operator Splitting for Fluid Dynamics in Complex Geometry (arxiv.org)

Pairwise Learning Increases Data Efficiency in Deep Neural Networks (arxiv.org)

Discount Yourself: Self-inclusive Freshness-weighted Gossip for Decentralised Learning under Client Outages (arxiv.org)

Efficient Edge-aware Attention Network for Graph Generation (arxiv.org)

When does Subgroup Balance Matter? Predicting Performance Sensitivity from Latent Separability (arxiv.org)

From Crawled to Published: A Publisher-Controlled Web Architecture for an AI-Native Internet (arxiv.org)

GraphCast-Mamba: Learning Non-Markovian Corrections for Weather Forecasting (arxiv.org)

ECG for laboratory values:A comprehensive physiological and technical analysis (arxiv.org)

Present but Hard to Read: Probing Latent Reasoning with Minimal Pairs (arxiv.org)

Model- and Layer-Level Closed-Loop Forecasting with SSM-Based Spiking Neural Networks (arxiv.org)

Geometry-Agnostic Data-Driven Discovery of Extreme Scenarios (arxiv.org)

LocaLeJEPA: Maximum Entropy Representations under Geometric Priors (arxiv.org)

LatticeMind: A Validation-Gated Agentic Framework for Density Functional Theory Workflows (arxiv.org)