AI News

LazyAgent: Demand-Driven Materialization and Physical Optimization of Agentic Programs (arxiv.org)

Enforcing Narrative Reliability and Epistemic Pacing in LLM-Driven Detective Games via Structured Knowledge Trees (arxiv.org)

Block-Level Weight-Space Structure Persists Under Post-Training: An Empirical Study Across LLM Families (arxiv.org)

Faithful Faithfulness Evaluations: Challenges & Pitfalls Learned from a Breast MRI Case Study (arxiv.org)

From Risk Scoring to Risk Allocation: A Density-Driven Framework for Diverse Monitoring in Multi-Agent Systems (arxiv.org)

Targeted Review for AI-Assisted Biodiversity Surveys: Active Continuous-Score Occupancy Modeling (arxiv.org)

From Experts to Sub-experts: Fine-grained Parameter-Efficient Fine-Tuning for MoE LLMs (arxiv.org)

An Exploratory Replica-Overlap Probe of the Grokking Transition (arxiv.org)

Generative Embodied Multiple Behavior Control Systems for Human-like Agents (arxiv.org)

Self-Organizing Agent Teams Learn to Reason Together (arxiv.org)

Terminal Shrinkage Averaging Reveals a Schedule-Estimator Interaction in LLM Pretraining (arxiv.org)

Mean Velocity Matching: Rethinking Generative Dynamics in Diffusion Models (arxiv.org)

PermuFormer: Multi-Task Pretraining for Permutation Representation in Algebraic Combinatorics (arxiv.org)

Lightweight Ranking Heads: Accelerating Multi-Task Experimentation in Production Recommender Systems (arxiv.org)

High-Order Liquid Evidence Modeling for Continuous and Subtle GNSS Spoofing Detection in Autonomous Driving (arxiv.org)

MSA-CITE: A Co-Adapted LoRA Specialist Ecology for Fixed-Budget Small-Model Inference (arxiv.org)

Predictive Uncertainty for Neural CAE Surrogates (arxiv.org)

Deep Reinforcement Learning on Item-Compatibility Graphs for One-Dimensional Bin Packing (arxiv.org)

Extending FunctionGemma for Practical On-Device Mobile Function Calling (arxiv.org)

Concept Drift from a Causal Perspective (arxiv.org)

MT-ProtBERT: Multi-task Learning ProtBERT for Intrinsically Disordered Proteins Classification with Scarce Data (arxiv.org)

Polyak-Type Extragradient Methods for Monotone Root-Finding Problems (arxiv.org)

LLM-based Conversational AI Knowledge Assistant for MyBuddy Humanoid Robot (arxiv.org)

AgentSTAR: Agentic Shape Tracking and Reconstruction from Monocular Videos (arxiv.org)

Harness-Zero: Harness Distillation via Agent-as-Harness (arxiv.org)

Emergent Collusion in Long-Horizon LLM Agent Interaction (arxiv.org)

Beyond Reconstruction Error: Analytical and Data-Driven Action Tokenization for Autoregressive Vision-Language-Action Models (arxiv.org)

Graded Representation Theory of Equivariant Neural Networks (arxiv.org)

Et Tu, Brute? Economic Misalignment in Personal AI Agents (arxiv.org)

Tutoring Large Language Models to be Domain-adaptive, Precise and Safe (arxiv.org)

How Strongly Should Task State Influence an LLM Agent? (arxiv.org)

Bridging the Data Gap: Digital Twin as a New Paradigm for AI-based Radio Sensing (arxiv.org)

Partially Observed Sparse Graphs: The Unknown Sampling Rate is a Tail Index (arxiv.org)

Beyond Imitation: Auditing the Recoverability of Reasoning in Distilled Models (arxiv.org)

From Research Frontier to Laboratory Bench: Design of a Four-Tier Experimental Teaching System for Multimodal Medical Image Intelligent Diagnosis (arxiv.org)

DVA-Neurons: Design and Verification of Adaptive LIF Neurons: From Single-Neuron Dynamics to Multi-Neuron Spiking Networks (arxiv.org)

ProcessLight: Process Supervision for Large Language Model Based Traffic Signal Control (arxiv.org)

LADDER: Graph-Guided Diffusion Language Models for Efficient Multi-Hop Reasoning (arxiv.org)

Brain-Token Learning: Microstate-Based Tokenization and Multi-Scale Interaction for Long-Horizon EEG Sequence Modeling (arxiv.org)

Error Bounds for Statistical Estimators in BTL Model with Parametric Multivariate Utility Functions (arxiv.org)

Correcting Within-Group Self-Selection Bias in Prioritized Replay (arxiv.org)

Direct Optimization of Generators for Search in Automated Theorem Proving (arxiv.org)

Recursive self-improvement of AI research agents (arxiv.org)

Generalized Deep Regression for Repeated Measurements (arxiv.org)

On the Gradient Heterogeneity Dynamics of Adversarially Robust Federated Regression (arxiv.org)

A Practical Recipe for Semi-Supervised Federated ASR: Online Pseudo-Labels with Server Update Stabilization (arxiv.org)

PP-Net: A Hybrid Physical-Prior Neural Network for Scattered Light Removal in Biomedical Images on Embedded Devices (arxiv.org)

On Basis Function Selection for Sparse Gaussian Process Regression (arxiv.org)

What Should a Self-Teacher See? Privileged Context Design for On-Policy Self-Distillation (arxiv.org)

Scalable Minimum-Volume Simplex Estimation with Non-asymptotic Analysis (arxiv.org)