AI News

MC-EvoBench: Auditing Experience-Driven Self-Improvement in Minecraft Agents (arxiv.org)

SpatialASM: Internalizing Active Mentalization for Spatial Reasoning (arxiv.org)

Task Context Shapes Component Roles in Language Models (arxiv.org)

Revisiting Tabular Deep Regression: Objectives and the Limits of Global Early Stopping (arxiv.org)

Hide and Seek: HIDE Concepts to SEEK explanations for black box models (arxiv.org)

DeltaLLM: Delta-Key Sparse Attention for Edge LLMs (arxiv.org)

Global Consistency Repair for Natural-Language Claim Sets (arxiv.org)

When Simulation Lies: A Sim-to-Real Benchmark and Domain-Randomized RL Recipe for Tool-Use Agents (arxiv.org)

Partially Informed Stochastic Optimization via Zeroth-Order Residual Tracking (arxiv.org)

Prediction Is Not a Sufficient Memory State (arxiv.org)

Designing Protein Structures from Natural Language (arxiv.org)

KL Before and After Mixing: A Round-Complexity Separation in Parallel Sampling (arxiv.org)

A Sampling Complexity-aware Framework for Discrete-time Fractional-Order Dynamical System Identification (arxiv.org)

Neuromorphic Satellite Intelligence: A Formal Framework for Event-Driven Architectures, Spike-Based Algorithms, and Quantitative Energy-Accuracy Trade-offs (arxiv.org)

TTS-Q: Perception-Guided Post-Training Quantization of Text-to-Speech Diffusion Transformers (arxiv.org)

Acoustic Biomarkers for Alzheimer's Disease Detection from Spontaneous Speech (arxiv.org)

Vocal Intensity Conversion with Continuous SPL Control via Adversarial Training (arxiv.org)

EarLog: Opportunistic Ear Physiology Sensing toward Personalized Hearing Risk Awareness (arxiv.org)

CORAL-KV: Compact Cross-Layer Prediction and Residual Coding for Autoregressive Video KV Caches (arxiv.org)

Looking Under the Streetlight: Evaluation in Generative Molecular Dynamics (arxiv.org)

MolP2SBench: Do LLMs Really Reason from Molecular Properties to Structure? (arxiv.org)

Electron-Density-Informed Molecular Property Prediction Using Surrogate Functionals (arxiv.org)

Structural and Representational Agency in LLM-Driven Timbre Generation and Interpolation (arxiv.org)

Fly-by-Code: Embodied Coding Agents for Aerial Manipulation with Active Visual and Physical Feedback (arxiv.org)

Toward Maximizing Algebraic Connectivity via Density-Dependent Backbone Construction and Reinforcement Learning Completion (arxiv.org)

Product Monte Carlo: Expectation Under a Product of Densities (arxiv.org)

Bridging Human and Model Attention: Explainability Analysis of CNN, Mamba, and ViT Architectures with Gaze-Based Validation (arxiv.org)

DREAM: Dynamic Resolution Assignment For Multimodal Multi-agent Debate (arxiv.org)

OpenCollab: A Multi-Agent Coding Framework with Programmable Collaboration and Controllable Runtime (arxiv.org)

Making LLMs Say What They Think: Measuring and Improving CoT-Interpretability Alignment (arxiv.org)

Scale and Selection: What Makes Automatic Harness Evolution Work for Visual-Interface Robot Agents (arxiv.org)

GOTT: Object-centric Dexterous Manipulation with a Reusable Cross-Embodiment Primitive (arxiv.org)

Solving VeriContest with a Lean-Backed Rust Verifier (arxiv.org)

AgroGround: Multi-Granularity Grounded Recognition in Agriculture (arxiv.org)

Video2World: Benchmarking Coding Agents for Interactive World Modeling from Embodied Videos (arxiv.org)

Green-Routed Neural Operators: Physics Determines Where the Network Reads (arxiv.org)

Decoupling Logic from Persona: Structural Immunity of Edge LLM Agents to Context Pollution (arxiv.org)

PHRBench: A Behavioral Evaluation of Post-Hallucination Reasoning in LLMs (arxiv.org)

VLA-FAIL: Efficient Task Failure Detection for Finetuned Vision-Language-Action Models (arxiv.org)

LLM4Trust: Exploring the Capabilities of Large Language Models for Trust Evaluation (arxiv.org)

Beyond In-Distribution Preservation: Recovering Generalization in Quantized VLAs via Vulnerability-Oriented Tuning (arxiv.org)

Beyond Prediction-Set Size: Quantifying Conformal Uncertainty through Credal Geometry (arxiv.org)

MetaJoin: Adaptive Join Selections for Heterogeneous Tables (arxiv.org)

Learning to accelerate Krasnosel'skii-Mann fixed-point iterations with guarantees (arxiv.org)

RPP: A Certified Poisoned-Sample Detection Framework for Backdoor Attacks under Dataset Imbalance (arxiv.org)

A Generalizable Machine Learning Framework for Half-Heusler Thermoelectrics (arxiv.org)

HumanX: Toward Agile and Generalizable Humanoid Interaction Skills from Human Videos (arxiv.org)

X Is Bad, Though (lexontech.org)

À Évry, la ligne de boissons des étudiants attend les industriels - La Clé Publique (laclepublique.fr)

Crest – an Arc-inspired open-source browser for Mac (crestbrowser.com)