AI News

One Size Does Not Fit All: Distribution-Guided Tensor-Wise Adaptive Sparsification for Model Merging (arxiv.org)

RCR-Router: Efficient Role-Aware Context Routing for Multi-Agent LLM Systems with Structured Memory (arxiv.org)

There are Levels to It: Red Teaming Dialogue With Hierarchical Reinforcement Learning (arxiv.org)

Unequal Voices: How LLMs Construct Constrained Queer Narratives (arxiv.org)

Physics-Informed Graph Neural Networks to Reconstruct Local Fields Considering Finite Strain Hyperelasticity (arxiv.org)

Scaling Vision Transformers Without Training: Prune-and-Merge Token Compression (arxiv.org)

From Individuals to Interactions: Benchmarking Gender Bias in Multimodal Large Language Models from the Lens of Social Relationship (arxiv.org)

RCStat: A Statistical Framework for using Relative Contextualization in Transformers (arxiv.org)

Co-Creative Learning via Metropolis-Hastings Interaction between Humans and AI (arxiv.org)

Simulating Student-Tutor Interactions with LLMs for Multilingual Math Feedback (arxiv.org)

SWEPatchTest: Evaluating Software Tests Beyond Reproduction (arxiv.org)

Benchmarks Are Not Atomic: Composition-Aware LLM Evaluation using BenchHub (arxiv.org)

Generative AI for Autonomous Driving: A Review (arxiv.org)

Graph Neural Network-Based Reinforcement Learning for Controlling Biological Networks - the GATTACA Framework (arxiv.org)

A Comprehensive Review on RNA Subcellular Localization Prediction (arxiv.org)

FASHION-RQ: Mining Hierarchical Semantic Structure for Automated Evaluation of Fashion Retrieval (arxiv.org)

ProtoArena: Can AI Agents Discover and Exploit Real-World Cross-Component Protocol Vulnerabilities? (arxiv.org)

A Tutorial on Quantum Amplitude Embedding with Data Re-Uploading (arxiv.org)

Agentic-MatriBO: Capability-Compiled Agentic Bayesian Optimization for Materials and Chemical Design (arxiv.org)

Diff Diffusion: Predicting Edit Paths with Diffusion Models for Efficient Code Editing (arxiv.org)

QK-Edit: Revisiting Attention-based Injection in MM-DiT for Image and Video Editing (arxiv.org)

Checking the Checkers: A Synthesis of Verifier Noise and Robustness in RLVR (arxiv.org)

Knowing the User Is Not Knowing What We Established: Grounding Awareness in Personal Agents (arxiv.org)

To Align or Not To Align: Check Your COMPASS Before You Train (arxiv.org)

Can an Agent Learn to Predict Its Own Failure? Outcome-Grounded Confidence Calibration in LLM Agents (arxiv.org)

Beyond Answer Voting: Fine-Grained Correction of LLM Reasoning through Graph Alignment across Sampled Traces (arxiv.org)

AgentStripes: An Evidence-Graded Maturity Model for AI Agents (arxiv.org)

Auditing Future-Memory Use in Point-in-Time Language Agents (arxiv.org)

Geometric Calibration of Supervised Fine-Tuning (arxiv.org)

Toward a Universal Characterization of Sudoku Puzzle Difficulty (arxiv.org)

Decomposability-Guided Cooperative Coevolution for Large-Scale Itinerary Planning (arxiv.org)

Agent-Environment Alignment via Automated Interface Generation (arxiv.org)

Compare to Correct: Pairwise-Guided Logit Adjustment for LLM Test-Time Scaling (arxiv.org)

LLM-BabyBench: Can Language Models Plan in Worlds They Can Simulate? (arxiv.org)

Physics: Benchmarking Foundation Models on University-Level Physics Problem Solving (arxiv.org)

Overcoming a Learning Bottleneck in Impartial Games: A Circuit Complexity Approach (arxiv.org)

Identification of Entailment and Contradiction Relations for Natural Language Argumentation: A Neurosymbolic Approach (arxiv.org)

What Matters in Quality-Aware VLA Pretraining: A Benchmark and Empirical Study (arxiv.org)

Tev: Scaling Long-Horizon Agentic Environments with Dense Progress Rewards (arxiv.org)

DoubtLess: Training-Free Contrastive Decoding for Efficient Reasoning (arxiv.org)

Hybrid Reasoning Agents (arxiv.org)

Decidable Reasoning About Time in Finite-Domain Situation Calculus Theories (arxiv.org)

Two Clocks: Why Larger Rollout Batches Learn Slower but Fail on Schedule in Reinforcement Learning with Verifiable Rewards (arxiv.org)

LIBERO-RIG: A Hardware-Setup Robustness Benchmark for Vision-Language-Action Models (arxiv.org)

Back-Translation Still Matters: Leveraging Monolingual Data for Low-Resource Machine Translation with LLMs (arxiv.org)

Optimizing Team Behavior via Extremum-seeking Control in Multi-agent Reinforcement Learning (arxiv.org)

ChunkAct: Action Chunking for Agentic Robotics (arxiv.org)

SEAR: Self-Evolving Agentic Robotics (arxiv.org)

Bytes to Threats: Byte-Sequence Transformer For Malware Detection (arxiv.org)

Where Do LLM Ensemble Gains Come From? Decomposing Candidate Value, Diversity, and Aggregation (arxiv.org)