AI News

Improving the Bounds on Approximating Weighted Maximin Share for Chores (arxiv.org)

TurnCost: Turns, Not Observations: Cache-Aware Loop Engineering for Coding Agents (arxiv.org)

scHFBiMamba: a bidirectional Mamba autoencoder for cardiac disease-state classification from single-nucleus RNA-seq data (arxiv.org)

Lifted State Policy Optimization: Internalizing Self-Correction through Transition Geometry (arxiv.org)

ShellElastic: Scalable Elastic Pretraining with Simultaneous Maximal Updates and Hyperparameter Transfer (arxiv.org)

Optimization for One, Feasibility for Most: Decoupling Optimization from Feasibility in Chance-Constrained Programs (arxiv.org)

Smart Alone, Blind Together: Why Accuracy Fails to CertifyTheory of Mind in Multi-Agent Teams (arxiv.org)

Revisiting Generative vs. Dense Retrieval: A Static Value Prior Reverses the Gap (arxiv.org)

Similar Models, Broken Odds: Where Luce's Axiom Fails in Chatbot Arena (arxiv.org)

RIECNN: Real-Time Image Enhanced CNN for Traffic Sign Recognition (arxiv.org)

To Defer or Not to Defer? Online Learning with LLMs and Congested Experts (arxiv.org)

UDHRBench: Benchmarking Human Rights in Large Language Models (arxiv.org)

Exploring Jailbreak Vulnerability Across Depth-Based Dynamic Inference in Large Language Models (arxiv.org)

PANINI-DR: Controlled Deep Research through Generative Semantic Workspaces (arxiv.org)

Framework for an AI-Ready ARPES Community Database (arxiv.org)

EvolvAttack: Backdooring self Evolving VLM Agents (arxiv.org)

Sequent-Prover: Training Agents for Formal, Checkable SMT-based Reasoning (arxiv.org)

LeHami: Symplectic Latent Dynamics for World Models (arxiv.org)

CTIA-ODE: Continuous-Time Inter-Variable Attention for Long-Term Time Series Forecasting (arxiv.org)

From Copying to Composition: How Language Models Develop In-Context Learning (arxiv.org)

Online Tokenization with Optimal Delay in Linear Time (arxiv.org)

GeoBridge: Evolving Transferable Decoder Topologies for Geometric Vision (arxiv.org)

North or South? Diagnosing LLMs' Understanding of North Korean Language, Society, and Knowledge (arxiv.org)

Posterior Flow Matching (arxiv.org)

Learning Who Benefits Most: Multi-Treatment Causal Ranking for Population Health Stratification (arxiv.org)

Reinforcement Learning via Advantage Guidance (arxiv.org)

VaryST: Generalizable Spatio-Temporal Forecasting Beyond the Closed-World Setting (arxiv.org)

When Does Depth Help? Understanding Layerwise Inference in Tabular Foundation Models (arxiv.org)

MAPLE: A Multi-Axis Parametric Knowledge Extraction Lens for LLM Evaluation (arxiv.org)

Analyzing and Mitigating Poisoning Attacks on Flow-Based Intrusion Detection (arxiv.org)

Adversarial Reinforcement Learning for Robust Planetary Landing (arxiv.org)

Rent-Seeking Language Models: Reward Models Set the Prices, Frequent Audits Curb the Rent (arxiv.org)

Memory-Efficient Subspace Optimization with Residual Correction: From Momentum SGD to Muon (arxiv.org)

Energy Consumption Analysis of Discrete Diffusion and Autoregressive Language Models (arxiv.org)

SPARS: A Stability-Plasticity-Aware Region Selection Framework for Continual Learning (arxiv.org)

When Non-Adversarial Noise Hurts Learning: Benchmarking Skill Self-Evolution in Agents (arxiv.org)

When Does the Dependency Graph Help Code-to-Architecture Mapping? (arxiv.org)

Occupancy Modeling via Drifting: One-Step Generation Enables Direct Policy Extraction (arxiv.org)

On the Limits of Representation-Based Test-Time Adaptation for Multimodal LLMs (arxiv.org)

Universal Sample-Based Embeddings of Higher-Order Objects (arxiv.org)

Seeing at the Edge: Energy Characterization of Vision-Language Model Inference on Edge Platforms (arxiv.org)

GREAT-ESDD: Graph Edge Attention Network for Effectively Steered Delivery Decisions (arxiv.org)

CESNET-TimeSeries26: A Multi-Year Network Traffic Time Series Dataset for Annual Seasonality, Concept Drift, and Long-Term Anomaly Detection (arxiv.org)

ASIWorld: Benchmarking Autonomous Agents in a Persistent Survival Sandbox (arxiv.org)

A Reference Architecture for Integrating Agentic AI into Enterprise Software: Responsibilities, Interactions, and Variability (arxiv.org)

Radioactive Feedback: Tracing Models Trained by Label-Only LLM Supervision (arxiv.org)

Continuous Risk-Based Scheduling for Food Safety Inspection: Integrating Historical Decay with Event-Driven Consumer Review Signals (arxiv.org)

EvoCowork: A Benchmark for Real-World Office Tasks (arxiv.org)

Pre-Registered Genesis Accounting for Recursive Self-Improvement: First Audited Operator-Genesis Events in a Physics-Discovery Runtime (arxiv.org)

PEAM: On-device Communication Agent for Personalized, Efficient and Autonomous Mobile Network Configuration (arxiv.org)