AI News

Think Wider: Mitigating Latent Rank Collapse in Implicit Chain-of-Thought Reasoning (arxiv.org)

Inferring Urban Mobility Interactions from Aggregated Dynamics (arxiv.org)

Robust Decentralized Personalized Federated Learning via Prediction-Constrained Neighborhood Collaboration (arxiv.org)

PCFlow: Physics-Conditioned Flow Matching for GPR B-Scan Image Synthesis (arxiv.org)

Constitutive State-Space Modeling of Path-Dependent Plasticity: A Resolution-Consistent and Parallelizable Computational Framework (arxiv.org)

Dense Structural Compression of Transformers via Gauge-Correct Channel Removal (arxiv.org)

Kolmogorov--Arnold stability for discontinuous functions (arxiv.org)

Parallelism Strategy Chaining for Fast Training Convergence (arxiv.org)

Robust Decentralized Federated Distillation via Multi-Modality Knowledge Collaboration (arxiv.org)

Protocol effects on feature-based hardware-Trojan detection across Trust-Hub families (arxiv.org)

FedRAW: Preserving Rare-Label Influence in Asynchronous Federated Learning (arxiv.org)

The Oversight Gap: What LLM Safety Monitors Miss, and Why It Is Not Capability (arxiv.org)

Stable-MM-R1: Anchoring Multimodal Reasoning Dynamics via Entropy-Guided Stratification (arxiv.org)

STAR-Pro: Stage-Wise Token Adaptive Reduction with Progressive Refinement for Efficient Large Vision-Language Models (arxiv.org)

SerenAI: State-transition system inspired by text-based world AI models (arxiv.org)

MARBO: Relational Belief Grounding for LLM Agents in Social Deduction Games (arxiv.org)

A Unified Policy Architecture (UPA): The Governance Kernel for Enterprise AI Operating Systems (arxiv.org)

Causal Attribution for Agentic Decisions: Estimators, Coupling, and a Traceability Specification (arxiv.org)

Personalizing LLM Agent Memory Using Biometrics (arxiv.org)

From Concentration to Differentiation and Back: Routing Effective Rank in MoE Reasoning Cohorts (arxiv.org)

REFINE: Trajectory Representation Learning via Closed-Loop Transcription -- Extended Version (arxiv.org)

Building Trustworthy Graph-Agentic RAG for Social Good: Architectures, Failure Propagation, and Assurance by Construction (arxiv.org)

AutoKD: Autonomous Knowledge Discovery (arxiv.org)

Predicting Wind Turbine Power Using Machine Learning and Weather Forecasts (arxiv.org)

SCIRIGOR:Evaluating Open-Ended Scientific Analysis Beyond Final Scores (arxiv.org)

Customer Relationship Intelligence: Integrating CRM and MDM for Enhanced Customer Engagement (arxiv.org)

MVFA: A Multi-View Text-Guided Multimodal Fusion LLM Adapter for Sentiment Analysis and Emotion Recognition (arxiv.org)

IIns-VAE+: A Robust Transfer Learning Framework for Environmental Identification in Wireless Sensing (arxiv.org)

Substrate-Portable Execution for Production LLM Workflows (arxiv.org)

CWF: A Collaborative Writing Framework for Personalized and Reliable Popular Science Writing (arxiv.org)

SAP: State-Guided Data Synthesis with Argument Provenance for Multi-Turn Tool Use (arxiv.org)

LayerRoute: Action-Conditioned Mixture-of-Layers Routing for Vision-Language-Action Policies (arxiv.org)

Generating Instance Generators in PDDL Planning (arxiv.org)

Explaining AI Agents Through Execution Traces (arxiv.org)

DAREBench: Deployment-Aware and Reliable Evaluation of Models as Agents (arxiv.org)

A Translational Note on AI Safety Evaluation (arxiv.org)

Generator-Independent Runtime Assurance under Partial Observation (arxiv.org)

Multi-Task Learning for Sparsely-Labeled Time Series: A Case Study on Cold-Hardiness Modeling (arxiv.org)

MOAE: Multi-Objective Agent Evolution with Pareto-Preserving Search (arxiv.org)

Beyond Final Decisions: A Process-Centric Benchmark for Transparent AI-Assisted Peer Review (arxiv.org)

Behavioral Grammar: Detecting Adaptive Malware via Tiny Language Model Priors and Second-Order Temporal Analysis (arxiv.org)

The End of AI Exponentiation: Fluttering Inside and Outside AI Bubble (arxiv.org)

Multimodal Resource-Exhaustion Attacks on Vision-Language Models via Joint Pixel-Prompt Optimization (arxiv.org)

AgentBrew: Offline Tool-Use Agent Learning from Raw Real-World Trajectories (arxiv.org)

The Emerging AI Paper-Review Arms Race: Adversarial Co-Evolution in Scholarly Publishing (arxiv.org)

It Is Not My Code Anymore (arxiv.org)

Aegix Pulse: A Traceable Three-Stage Architecture for Personalized Content Generation and Context-Preserving Revision (arxiv.org)

Norms at a Price: Why RL-Based Alignment Can Promise Conditional Compliance at Best (arxiv.org)

AgentIdeaBench: Benchmarking Scientific Ideation in the Agent Era (arxiv.org)

FinCUABuild: Can Agents Build Reliable Benchmarks for Dynamic Financial Computer Use? (arxiv.org)