AI News

Combinatorial Characterizations of the Learnability of Conditional Distributions (arxiv.org)

Spatial Heuristic Strategy based Diffusion Reinforcement Learning for Online 3D Bin Packing (arxiv.org)

Training an MoE LLM in MXFP4: How We Closed the Gap to BF16 (arxiv.org)

Federated Nonparametric Regression with Heterogeneous Clients (arxiv.org)

Exact Global Minimax Risk near Nonidentification (arxiv.org)

Score-Preserving Continual Lasso under Task Shift (arxiv.org)

Data-Efficient Backdoor Attacks on Code Models via Trigger-Neutralized Counterfactual Scoring (arxiv.org)

Interpretable Singing Emotion Recognition Using Acoustic and Vocal Technique Features (arxiv.org)

Decoupling Patient Personas and Clinical Workflows: A Dual-Agent Approach to Realistic Medical Dialogue Generation (arxiv.org)

DriveMind: Mind-Evolving Belief Tracking for Closed-Loop Autonomous Driving (arxiv.org)

Learning Nonlinear Waveform Fusion Across Scales for Integrated Sensing and Communication (arxiv.org)

SWE-Shift: Evaluating Coding Agents in Evolving Workspaces (arxiv.org)

MPQ: A Message-Passing View of Post-Training Quantization (arxiv.org)

Not All Information Is Worth Asking For: Decision-Informed Information Seeking with LLMs (arxiv.org)

Correct Answers, Closed Gates: SolveBind for GUI Agents at the Verification Barrier (arxiv.org)

Implicit Neural Representations for Variational Problems on Graphons (arxiv.org)

EditJev: System One Routing for Image Editing Models (arxiv.org)

Tool-Calling LLM Agents Under Realistic MCP Faults: Scale Does Not Purchase Fault Tolerance (arxiv.org)

Behaviour analysis for large language model-driven agents (arxiv.org)

SP-Guide: Enhancing Aerial Vision-Language Navigation with Spatial Priors (arxiv.org)

StarRTT: An End-to-End Real-Time Framework for Blind Zone-Free RTT Prediction in Starlink LEO Satellite Networks (arxiv.org)

Async-RBench: Benchmarking Asynchronous Task Execution in LLM Agents (arxiv.org)

A Mathematical Theory of the Jacobian Lens (arxiv.org)

What Transfers Across Protein Pockets? Experience Reuse in Molecular Optimization for Agentic Structure-Based Drug Design (arxiv.org)

Power- and Fragmentation-aware Online Scheduling for GPU Datacenters (arxiv.org)

Tradeoff-Aware Territory Design: A Multi-Objective Decision System for Enterprise Customer-to-Seller Allocation (arxiv.org)

HarnessMM: Recursive Self-Improvement for Harness Token Efficiency (arxiv.org)

Analysing Austrian Parliamentary Speeches by Predicting Party Affiliations (arxiv.org)

Enhancing Table Recognition with Vision LLMs: A Benchmark and Neighbor-Guided Toolchain Reasoner (arxiv.org)

Testing Transformer Learnability on the Arithmetic Sequence of Rooted Trees (arxiv.org)

EvoSkill v2: Fully Agentic Skill Evolution from Benchmark Adaptation to a Stateful Evolver (arxiv.org)

REACT: Representational Early Analysis and Control of Chain-of-Thought Trajectories (arxiv.org)

Advancing SVD-based LLM Compression via Layer-Wise Error Model Search (arxiv.org)

Learning Structured Sparsity for Large Models: Bridging Model Compressibility and Efficient Inference (arxiv.org)

Understanding Communication Failures in Multi-Agent LLM Systems (arxiv.org)

Adaptive World-Models for Anomaly Recognition (AWARE): A Framework for Robot Dynamics Prediction and Introspection (arxiv.org)

Multiplicative vs. Additive State Coupling in Learned Optimizer Schedules (arxiv.org)

AlphaMemo: Structured Search-Process Memory for Self-Evolving Alpha Mining Agents (arxiv.org)

Validating the Comparison, Not Just the Scorer: Why Overall Accuracy Can Mislead in Interactive LLM Evaluation (arxiv.org)

Spectra: Multi-Scale Implicit Decomposition for Deepfake Detection (arxiv.org)

LLaCO: Continuous Diffusion for Large Language Models with Competition-Aware Semantic Scheduling (arxiv.org)

CASE: Confusion-Aware Skill Evolution for Frozen Multimodal Emotion Models (arxiv.org)

TGR-EvoDAG: A DAG-based Multi-Agent Code Evolution Framework for Generative Recommendation on Tencent (arxiv.org)

ReMedI: Inference-Time Refinement and Self-Distillation for Controllable Molecular Optimization (arxiv.org)

From Deliberately Constructed Systemic Representations of the World of Activity to Dialogue: A Five-Level Communication Architecture for Strong Language-Based ASI (arxiv.org)

NgramRL: Accelerating RL Post-Training for Structured Generation via Rollout-Aware N-gram Speculative Decoding (arxiv.org)

Bridging Exploration and Persona: Information-Directed Dueling for Test-Time Agent Personalization (arxiv.org)

Effectiveness of Curriculum Learning Depends on Reward Sparsity and Competing Optima (arxiv.org)

IGG: A Benchmark for Interactive GUI Grounding under Visibility Constraints (arxiv.org)

LGMT: A Tool for Testing the Logical Reasoning Reliability of LLMs (arxiv.org)