AI News

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

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

Beyond Copying: Lifelong Smart Contract Security Task Construction via Protocol Mutation (arxiv.org)

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

4DMemory: Benchmarking Long-Horizon Spatiotemporal Memory with Harness-Augmented Agents (arxiv.org)

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

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

Closing the Modality Gap via Simplex-Constrained Representations (arxiv.org)

PASTA: Prompt-Aware Annealed Diffusion Sampling (arxiv.org)

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

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

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

RobustRLlib: A Unified Library and Benchmark for Robust Reinforcement Learning Algorithms (arxiv.org)

Collaborative In-Context Learning for Indoor Radio Map Estimation (arxiv.org)

SemantEq: Faithful Bidirectional Equivalence of Formal Statements (arxiv.org)

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

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

Principia: Test-Time Hypotheses to Principle Construction for Scientific Discovery (arxiv.org)

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

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

Federated Nonparametric Regression with Heterogeneous Clients (arxiv.org)

DIGOM: Disentangled Interaction Graphs for Outcome prediction in Multimodal healthcare (arxiv.org)

Symbolic Guidance for LLM Agents in Distributed Multiagent Coordination (arxiv.org)

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

Attractor Domain Theory: A Mathematical Framework for Cardiovascular Attractor Analysis with Wearable Photoplethysmography (PPG) Validation (arxiv.org)

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

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

MedInsightBench: Evaluating Medical Analytics Agents Through Multi-Step Insight Discovery in Multimodal Medical Data (arxiv.org)

ArtifactArena: Evaluating Models by What They Build in the Physical World (arxiv.org)

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

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

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

SceneBloom: Agentic Indoor Environment Generation with Realistic and Diverse Layouts for Embodied Learning (arxiv.org)

Rethinking LoRA Initialization for Robust Asymmetric Learning Rates (arxiv.org)

MEDMKG: Benchmarking Medical Knowledge Exploitation with Multimodal Knowledge Graph (arxiv.org)

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

Cycle-of-Thought: Self-Evaluative Iterative Reasoning for Table Visual Question Answering in Document Images (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)

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

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

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

AutoMIL: Automated Multiple Instance Learning for Whole-Slide Image Classification (arxiv.org)

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

The LLENS Platform for On-Premises Large Language Model Agents in a Japanese University Hospital: Development and Evaluation Study (arxiv.org)

Mining the Design Space of Multi-Agent Workflows (arxiv.org)

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

From Target Selection to Condition Reuse: Auditing Residual Interventions in Tool-Use Models (arxiv.org)

TENET: Time-point Encoding Network for Multivariate Time Series Anomaly Detection (arxiv.org)

Can Test-Time Memory Evolution Outperform Test-Time Training for Scientific Discovery? (arxiv.org)

An Agentic Approach with Evidence-Grounded Verification for Multimodal Emotion Reasoning in Conversation (arxiv.org)