AI News

AgentTrace: Execution-Level Fingerprinting of Black-Box LLM Agents (arxiv.org)

Concept Acupuncturist: An Efficient and Precise Concept Erasure Method for Diffusion Models via Sparse Autoencoders (arxiv.org)

Guided Zero-Shot Anomaly Explanation for Anonymized Crowd Video Monitoring: Context, Categorization, and Spatial Localization (arxiv.org)

HyScale: Scaling Long-Context Training for Learned Sparse Attention (arxiv.org)

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

OpenHome: From Perception to Insight in Smart-Home Camera Videos (arxiv.org)

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

Agentic AI Defense: Decode-Time Provenance Binding for MCP (arxiv.org)

RobustNCL: Robustifying Neuromorphic Continual Learning under Transient Faults with Lightweight Detect-and-Cure Strategy (arxiv.org)

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

Knowing When to Stop: The Evidence-Action Gap in Coding Agents (arxiv.org)

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

Kolibri: A Sovereign European Model on the Pareto Frontier (arxiv.org)

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

When Agents Fail: Model or Harness--Who's to Blame? (arxiv.org)

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

Beyond Post-Hoc Rewards: Privileged Reasoning Exploration for Multimodal Embeddings (arxiv.org)

ALLMTS: Automated LLM-based Medical Triage System (arxiv.org)

POSEIDON: Scaling Training Environments for Scientific Agents from Code Repositories (arxiv.org)

When to Stop Penalizing Actions? Martingale GRPO for VLA Reinforcement Learning (arxiv.org)

CauProphet: Teaching Agents to Develop Causal Cognition like Humans (arxiv.org)

DealsFlow: An Evidence-Gated Agentic Harness for Venture Capital Deal Triage (arxiv.org)

Do AI-for-Science Agents Preserve Model Safety? (arxiv.org)

LeafSkill: Efficient Leaf-Level Action Grounding for Recursive LLM Agents (arxiv.org)

PPO vs. SAC: A Comparative Study for Cooperative Motion Control of Dual-Arm Space Robots (arxiv.org)

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

Batched NN 4D-Var: a performant hybrid model framework for the Integrated Forecasting System (arxiv.org)

Acquisition for Spherical-Linear Bayesian Optimization on the Cube (arxiv.org)

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

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

Federated Nonparametric Regression with Heterogeneous Clients (arxiv.org)

Early learning in large residual networks (arxiv.org)

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

Autoregressive Physics-Informed Neural Operator Surrogates and Joint 4-D Seismic-Production Bayesian History Matching (arxiv.org)

Sparse Bayesian Neural Networks via Sticky Piecewise Deterministic Markov Processes (arxiv.org)

GeoDiv KAN Solver for Time Fractional Navier Stokes Equations on Complex Domains to Realistic CFD Reconstruction (arxiv.org)

Empirical concentration for chains with infinite memory (arxiv.org)

Tokamak-SI: Interpreting Tokamak Discharges with Labeled Catalogs (arxiv.org)

Adaptive Control of Scientific Instruments with Nonstationary Responses through Multimodal Agents (arxiv.org)

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

ACVA-Bench: Beyond Prompt Following in Abstract Concept Visualization Ability for Text-to-Image Models (arxiv.org)

What Open-Source Projects Tell Their Coding Agents: A Census of AGENTS.md and CLAUDE.md (arxiv.org)

Exact Global Minimax Risk near Nonidentification (arxiv.org)

Conditional Flow Models for Offline Maximum Entropy Reinforcement Learning (arxiv.org)

DeepOpen: Decision-Oriented Intent Classification on BANKING77 and CLINC150 (arxiv.org)

Yalda-Embedding: Open Development of a Competitive Persian Retrieval Model (arxiv.org)

WiMoT: Replay-Free Continual Wi-Fi Sensing via Moment Transport (arxiv.org)

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

ActionRAG: Evidence-Refining Graph Diffusion for Long-Term Memory in LLM Agents (arxiv.org)

Lifelong Context Adaptation through Bidirectional Memory-Skill Learning (arxiv.org)