AI News

Mechanical Field Networks: Structured Neural Dynamics for Multivariate Systems (arxiv.org)

Physics-informed generative AI for semiconductor manufacturing: Enforcing hard physical constraints in generative models by construction (arxiv.org)

Kuramoto Attention: Synchronizing Self-Attention on the Torus (arxiv.org)

ProHiFlo: Hierarchical Flow Matching with Functional Guidance for De Novo Protein Generation (arxiv.org)

Few-Shot Resampling for Scalable Statistically-Sound Data Mining (arxiv.org)

RoVE: Rotary Value Embeddings Attention for Relative Position-dependent Value Pathways (arxiv.org)

To Intervene or Not: Guiding Inference-time Alignment with Probabilistic Model Blending (arxiv.org)

Restless bandits with imperfect binary feedback: PCL-indexability analysis and computation (arxiv.org)

WorldReasoner: Evaluating Whether Language Model Agents Forecast Events with Valid Reasoning (arxiv.org)

Sparsified Kolmogorov-Arnold Networks for Interpretable Quantum State Tomography (arxiv.org)

TextHOI-3D: Text-to-3D Hand-Object Interaction via Discrete Multi-View Generation and Joint Mesh Optimization (arxiv.org)

Multimodal Ordinal Modeling of Alzheimer's Disease Severity Using Structural MRI and Clinical Data (arxiv.org)

Blind Dexterous Grasping via Real2Sim2Real Tactile Policy Learning (arxiv.org)

Fast Speech Foundation Model Distillation Using Interleaved Stacking (arxiv.org)

Automated Creativity Evaluation of Language Models Across Open-Ended Tasks (arxiv.org)

AnchorEdit: Maintaining Temporal Consistency in Multi-turn Image Editing via Causal Memory (arxiv.org)

From Prompts to Tokens: Internalizing Causal Supervision in Vision-Language Model for Multi-Image Causal Reasoning (arxiv.org)

Hey Chat, Can You Teach Me? Structuring Socratic Dialogue for Human Learning in the Wild (arxiv.org)

Multi-View In-Cabin Monitoring System for Public Transport Vehicles (arxiv.org)

ICA Lens: Interpreting Language Models Without Training Another Dictionary (arxiv.org)

Ouroboros-Spatial: Closing the Data-Model Loop for Spatial Reasoning (arxiv.org)

Substrate Asymmetry in User-Side Memory: A Diagnostic Framework (arxiv.org)

MedCTA: A Benchmark for Clinical Tool Agents (arxiv.org)

T2S: A Rehearsal-Based Approach for Extraction-Resistant Model Watermarking (arxiv.org)

Noise-Aware Framework for Correcting Corrupted Labels (arxiv.org)

Layer-Isolated Evaluation: Gating the Deterministic Scaffold of a Production LLM Agent with a No-LLM, Regression-Locked Test Harness (arxiv.org)

Reason, Then Re-reason: Cross-view Revisiting Improves Spatial Reasoning (arxiv.org)

Can Open-Source LLM Agents Replace Static Application Security Testing Tools? An Empirical Assessment (arxiv.org)

Runtime Skill Audit: Targeted Runtime Probing for Agent Skill Security (arxiv.org)

ARGUS: Stacked Multi-View Identity Mosaic Injection for Subject-Preserving Video Generation (arxiv.org)

Sparse probes and murky physics: a case study of interpretability challenges in a foundation model for continuum dynamics (arxiv.org)

TAROT: Task-Adaptive Refinement of LLM-prior Graphs for Few-shot Tabular Learning (arxiv.org)

Are LLMs Bad at Moral Reasoning? (arxiv.org)

Sovereign Assurance Boundary: Certificate-Bound Admission for Agentic Infrastructure (arxiv.org)

LUCID: Learning Embodiment-Agnostic Intent Models from Unstructured Human Videos for Scalable Dexterous Robot Skill Acquisition (arxiv.org)

When Context Returns: Toward Robust Internalization in On-Policy Distillation (arxiv.org)

Information-Theoretic Decomposition for Multimodal Interaction Learning (arxiv.org)

LakeFM: Toward a Foundation Model for Aquatic Ecosystems Using Irregular Multivariate Multi-depth Time Series Data (arxiv.org)

Physics-Distilled Neural Network enabled by Large Language Models for Manufacturing Process-Property Predictive Modeling (arxiv.org)

Model-Based and Data-Driven Hierarchical Control and Topology Co-Design for Robust Networked Systems (arxiv.org)

AVIS: Adaptive Test-Time Scaling for Vision-Language Models (arxiv.org)

ConsistencyPlanner: Real-time Planning with Fast-Sampling Consistency Models (arxiv.org)

LLMs+Graphs: Toward Graph-Native, Synergistic AI Systems (arxiv.org)

End-to-End Machine Learning for Depressive State Classification via EEG and fNIRS (arxiv.org)

Pretrained self-supervised speech models can recognize unseen consonants (arxiv.org)

AI Researchers Must Help Lead Arms Control to Mitigate Military AI Risks (arxiv.org)

ISE: An Execution-Grounded Recipe for Multi-Turn OS-Agent Trajectories (arxiv.org)

Dual-Stance Evaluation of Sycophancy: The Structure of Agreement and the Limits of Intervention (arxiv.org)

SirenFNO: Efficient and Full Frequency Learning of Fourier Neural Operators (arxiv.org)

On the Study of Biometric Spoofing Detection using Deep Learning (arxiv.org)