AI News

SciTrue: Reliable Scientific Claim Validation with Frontier and Open Language Models at the NTCIR SciClaimEval Task (arxiv.org)

DramaChain Bench: An End-to-End Benchmark for Short-Drama Generation (arxiv.org)

REVISE: Validity-Guided Recovery for Online Revisions in Agent Workflows (arxiv.org)

Control-Data Flow Separation: Stable Prompt Optimization in Multi-Agent LLMs (arxiv.org)

Socrates went Nuclear: Comparing Interaction Strategies for AI systems in a Learning Context using Brain Sensing (arxiv.org)

Escaping Redundant Reasoning: Structure-Aware Search for Inference-Time LLMs (arxiv.org)

ContextPipe: Database-Inspired Context Assembly for Long-Horizon Agents (arxiv.org)

Same Request, Different Boundary: Evaluating Cybersecurity Assistance across Conversational Contexts (arxiv.org)

Consistency Without Alignment: Item-Sensitive Language Models Indistinguishable From Random (arxiv.org)

Residual Sparsification via Output Importance for Compressing Mixture-of-Experts LLMs (arxiv.org)

Feedback-Assisted Trust Propagation over Document Relation Graphs for Retrieval-Augmented Generation (arxiv.org)

When the Algorithm Becomes the Brand Crisis: A Sociotechnical Theory of Distributed Responsibility and Accountable Transparency (arxiv.org)

CoVer: Conflict-Aware Claim Verification (arxiv.org)

Wave Function Backpropagation with Explicit Temporal-Interval Dynamics (arxiv.org)

Causal Evidentiary Governance for High-Risk Machine Learning Systems (arxiv.org)

FedReview: Review and Dispose Poisoned Updates without Validation Datasets or Historic Knowledge (arxiv.org)

HiveTraceGuard-Pro: A Compact Generative Guardrail for Prompt Injection, Jailbreaks, and Adversarial Obfuscation (arxiv.org)

Lagged Coupling: Internal Representations Become Readable Before They Become Causal (arxiv.org)

trajectory-judge: What Outcome-Only LLM Judges Miss on Agent Trajectories (arxiv.org)

ToSCA: Leveraging Hierarchical Reinforcement Learning on Temporal and Strategic Abstractions of Conversational Agents (arxiv.org)

MRMAD: A Multi-Round Multi-Audio Benchmark for Evaluating Acoustic Degradation Perception in Large Audio-Language Models (arxiv.org)

Freeze, Diffuse, Decode: Geometry-Aware Adaptation of Pretrained Transformer Embeddings for Antimicrobial Peptide Design (arxiv.org)

Scaling Near-Optimal SFT-RL Annotation Budget Allocation from Small to Large LLMs (arxiv.org)

Facet-0: A Robotic Foundation Model for Contact-Rich Precise Manipulation (arxiv.org)

Beyond Scores: Understanding LLM-as-a-Judge Mechanisms in Summarization Evaluation (arxiv.org)

Efficient Learning of Balanced Signed Graphs via Sparse Linear Programming (arxiv.org)

Towards Provable and Scalable Training of Quantized Neural Networks with Ising Optimization (arxiv.org)

Advantage Weighted Matching: Aligning RL with Pretraining in Diffusion Models (arxiv.org)

Hints Help But Do They Teach? Evaluating Skills Transfer in Code Generation (arxiv.org)

StainPresetNet: Stain Preset Network for Fast Multi-to-Multi Stain Normalization (arxiv.org)

Athena: Vulnerability-Affected Library Identification via Knowledge Graph Completion (arxiv.org)

Towards AI-Assisted Clinical Trial Matching: Practical Considerations, Multicenter Evaluation, and Real-World Deployment (arxiv.org)

Autonomous discovery of new structure-plausibility laws for explainable and rapid crystal diagnosis and screening (arxiv.org)

Who Judges the Judges? A Chinese Safety QA Benchmark for Evaluating LLM Responses and Safety Judges (arxiv.org)

Mechanism Design for Alignment and Control (arxiv.org)

Position Matters: Feature Inversion Attacks in ViT Split Inference with Token Reduction and Shuffling (arxiv.org)

MutMem-V2: Cryptographically Authorized Mutation in Persistent Agent Memory Portable Verification and Reproducible Evidence (arxiv.org)

One Prompt Is Enough: Watermark Laundering Through Foundation Image Models (arxiv.org)

TimeSteer: Inference-Time Speech Scheduling in Joint Audio-Visual Diffusion Models (arxiv.org)

Some Emotions Run Deeper: Layer-wise Probing and Causal Intervention in Large Language Models (arxiv.org)

EmbodiedSkills: A Unified Framework for Orchestrating, Training, and Deploying VLA Agents (arxiv.org)

Iterative GRPO: Batch-Online Policy Iteration for Multi-Turn RL via Single-Turn RLHF (arxiv.org)

HiLRP: Toward One Trustworthy Explanation for Vision Transformer: Conservation-Valid Attribution via Attention Primitives (arxiv.org)

Probing Factual Knowledge Transfer with Training Data Interventions (arxiv.org)

Scalable Rao-Blackwellized Online Planning for High-Dimensional POMDPs (arxiv.org)

CHARM: Character Hallucination for Multicultural Role Play Benchmark (arxiv.org)

Stress Testing Unlearning Algorithms (arxiv.org)

PopPert: Population-level Joint-Distribution Modeling for Single-Cell Perturbation Prediction (arxiv.org)

Evaluating Multimodal LLMs as Generalist Vision-Language-Action Agents for Drone Control: Commanding, Approaching, Tracking and Searching (arxiv.org)

When Safety Routing Breaks: Understanding Alignment Fragility under Benign Fine-Tuning (arxiv.org)