AI News

A Three-Tier Persona Vector for Controllable User Simulation in Agentic Evaluation (arxiv.org)

Graph-Based Personalized Memory for LLM Agents: Representation, Evolution, Retrieval, and Evaluation (arxiv.org)

CLAMP: Constrained Decoding for Vision-Language Embodied Planning (arxiv.org)

GoAnt: Quality-Diversity Multi-Agent Search for Alpha Factor Discovery in Market Microstructure Data (arxiv.org)

Robots Influencing Humans to Reveal their Goals during Collaboration and Competition (arxiv.org)

It's All in the Way You Say It: The Role of Information Representation in LLM-Based Glycemic-Event Prediction (arxiv.org)

Closing the Consistency Gap: Self-Evolving Agents That Learn to Stay on Course (arxiv.org)

API Benchmark Scores Do Not Reliably Transfer to Chatbot Interfaces (arxiv.org)

SkillAdam: Stable and Efficient Skill Evolution for Agents (arxiv.org)

PlannerForge: LLM Agents for Scenario-Based Testing of Motion Planners in Autonomous Driving (arxiv.org)

Good Pretraining, Bad SFT: Checkpoint Quality Across the Training Stack (arxiv.org)

Deposon: An Auditable, Conservation-Guaranteed, Game-Theoretically Tested Scattering Layer over LLM Reasoning Paths (arxiv.org)

Answer-Distribution Trajectories: A Stochastic-Dynamics View of LLM Reasoning (arxiv.org)

Time-Varying Data as Sheaves: an Invitation to Narratives (arxiv.org)

Everything in Moderation: Per-Domain Coverage Optima and Alignment-Resistant Domain Gaps in Multi-Domain Mid-Training (arxiv.org)

The Surprising Effectiveness of Approximate Value Iteration in Self-Play (arxiv.org)

SAEScientist-Bench: Can AI Agents Conduct Autonomous SAE Interpretability Research? (arxiv.org)

MeClear: Cooperative Game-Theoretic Attribution and Risk-Aware Memory Clearance for Long-Horizon LLM Agents (arxiv.org)

A Generalization of Amari's Bayesian Duality (arxiv.org)

ExecCritic: Learn to Test, Test to Improve for Coding Agents (arxiv.org)

Co-Evolving Harnesses and Models: On-Policy Correction Helps Weaker Models Catch Up Where Imitation Fails (arxiv.org)

A Data-Driven Framework for Identifying and Prioritizing RPA Opportunities in Healthcare Processes (arxiv.org)

Data-driven rational function neural networks: a new method for generating analytical models of rock physics (arxiv.org)

Emergent Goal-Directed Attention in Large Vision-Language Models (arxiv.org)

CrossModalQA: A Cross-modal and Multi-hop Benchmark for Multimodal Retrieval-augmented Generation (arxiv.org)

Situation Awareness for Intelligent Data Distribution in Connected Vehicles (arxiv.org)

Diffusion models for eye-gaze trajectory generation using position and velocity representations (arxiv.org)

DART: A DAG-Based Reputation and Incentive Framework via Blockchain-Enabled Governance for Trustworthy LLM Multi-Agent Collaboration (arxiv.org)

When Agent Governance Helps (arxiv.org)

Beyond the Verdict: Evidence-Aligned Evaluation of Visual Prompt-Injection Guardrails (arxiv.org)

Dual-Latent Memory Routing for Vision-Language Reasoning (arxiv.org)

Knowing When Not to Answer: Abstention and Refusal Reasoning in Vision--Language Models (arxiv.org)

Subject-Relative Micro-Motion and Sleep Dynamics for Near-Infrared Video Sleep Staging (arxiv.org)

An Agent Model Abstraction for Human-AI Teaming Cognitive Coupling (arxiv.org)

ViT3Flow: A Test-Time Training Transformer MeanFlow for Postoperative Radiograph Synthesis in Scoliosis (arxiv.org)

WolfSociety: Understanding Collective Risk from Harmful-Agent Scaling in Financial Agent Societies (arxiv.org)

LEBGen: An LLM-Enhanced Bayesian Network Framework for Few-Shot Travel Survey Data Generation (arxiv.org)

Learning Kernels by Alignment for Multiclass Bayes Classification (arxiv.org)

How Does Parameter Pruning Reshape DNN Representations? An Interaction-Driven Exploration (arxiv.org)

Inducing Emergent Misalignment from Reward Hacks with Iterative DPO (arxiv.org)

Second-Order Smooth Planning with Optimal-Transport Bellman Smoothing (arxiv.org)

Power Mean Estimation in Stochastic Continuous Monte Carlo Tree Search (arxiv.org)

Structural Entropy-Driven Graph Diffusion Generation for One-Shot Federated Graph Learning (arxiv.org)

Role-Specific Predictive Geometries for Nonstationary Multivariate Graph-Signal Forecasting (arxiv.org)

Not Just Oversmoothing: Detecting the Echo Chamber Effect in Graph Neural Networks (arxiv.org)

Hidden in Plain Sight: The Overlooked Significance of Canonical Elements for Extreme LLM Sparsity (arxiv.org)

Layer-Wise Gate-Controlled Prompt Truncation in a Multimodal Chest X-Ray Classifier (arxiv.org)

Deep Barycentric Regression for Optimal Transport Map Estimation and its Statistical Optimality (arxiv.org)

A Statistical and Machine Learning Framework for Quantifying Offensive Impact in Professional Box Lacrosse (arxiv.org)

SwiftExplorer: Training-free Diffusion Model Alignment with Swift Diversity Exploration (arxiv.org)