AI News

Cortex and subcortex play distinct roles over learning when cortical memory is limited (arxiv.org)

Information-Theoretic Lower Bounds for Bit-Constrained Stochastic Optimization via a Reduction to Compressed Gaussian Mean Estimation (arxiv.org)

ViBE: Co-Optimizing Workload Skew and Hardware Variability for MoE Serving (arxiv.org)

Scaling Parallel Sequence Models to Foundation-Scale Vision Encoders (arxiv.org)

Benchmarking Security Risk Detection and Verification in Open Agentic Skill Ecosystems (arxiv.org)

Detection vs. Execution: Single-Bucket Probes Miss Half the Mamba-2 State Sink (arxiv.org)

CV-Arena: An Open Benchmark for Instructional Computer Vision Problem Solving with Human-AI Collaborative Preferences (arxiv.org)

Lodestar: An Online-Learning LLM Inference Router (arxiv.org)

SS-ZKR: Spatial-Semantic Zero-Knowledge Routing for Privacy-Preserving Multi-Agent Collaboration (arxiv.org)

An Open-Source Benchmark and Baseline for Multi-temporal Referring Segmentation (arxiv.org)

FVSpec: Real-World Property-Based Tests as Lean Challenges (arxiv.org)

Cross-Axis Feature Fusion with Joint-Wise Motion Difference Prediction for Text-Based 3D Human Motion Editing (arxiv.org)

Internalize the Temperature: On-Policy Self-Distillation as Policy Reheater for Reinforcement Learning (arxiv.org)

Benchmark Dataset for Catalysis on 2D MXenes (arxiv.org)

Quality-Diversity Evolution for Discovering Diverse Vulnerabilities in LLM Safety (arxiv.org)

Generative Diffusion Priors for 3D Mapping of the Dark Universe (arxiv.org)

AI-IoT-Robotics Integration: Survey of Frameworks, Emerging Trends, and the Path Toward Connected Robotics (arxiv.org)

PolySpeech-100: A Large-Scale Benchmark for Speech Understanding Across 100+ Languages and Dialects (arxiv.org)

Hybrid Verified Decoding: Learning to Allocate Verification in Speculative Decoding (arxiv.org)

ProductWebGen: Benchmarking Multimodal Product Webpage Generation (arxiv.org)

Data Collection for Training Quality-Control AI in Carpet Manufacturing (arxiv.org)

DSL-LLaDA: Scaling Continuous Denoising to 8B Masked Diffusion LMs (arxiv.org)

Cross-Generational Transfer of Adversarial Attacks Reveals Non-Monotonic Safety Alignment in LLMs (arxiv.org)

Coarse-to-Fine Compositional Diffusion for Long-Horizon Planning (arxiv.org)

Sympatheia: Emotionally Adaptive Voice Assistant with Continuous Affect Conditioning (arxiv.org)

Temporally-Aligned Evaluation for Audio-Driven Talking Head Generation (arxiv.org)

3DCodeBench: Benchmarking Agentic Procedural 3D Modeling Via Code (arxiv.org)

MENTIS: What Belief Changes Under Alignment? Measuring Multi-Scale Latent Torsion in Language Models (arxiv.org)

Beyond Task Success: Behavioral and Representational Diagnostics for WAM and VLA (arxiv.org)

Implicit Drifting Policy: One-Step Action Generation via Conditional Expert Geometry (arxiv.org)

MiCU: End-to-End Smart Home Command Understanding with Large Language Model (arxiv.org)

AMP: A Vendor-Neutral Wire Format for Agent Memory Operations (arxiv.org)

ASE-26: a curriculum for agentic software engineering as a discipline (arxiv.org)

AI From the Margins (AIM): Rethinking Participatory AI Design Through the Lived Experience of Minoritized Communities (arxiv.org)

CA-BED: Conversation-Aware Bayesian Experimental Design (arxiv.org)

Topological Ignorability for Structural Causal Effects Beyond Means (arxiv.org)

Demystifying Multimodal Biomolecular Co-design With Intrinsic Geodesic Coupling (arxiv.org)

Meta-Black-Box Optimization with Ensemble Surrogate Modeling for Robustness-Accuracy Trade-off within SAEA (arxiv.org)

A Lightweight Hybrid MLP-Based Framework for Real-Time Phishing URL Detection Using Structural URL Features (arxiv.org)

Tiny Recursive Models for Solving the J2-Perturbed Lambert Problem (arxiv.org)

Reason, Retrieve, Re-rank: A Zero-Shot Reasoning-Aware Framework for Composed Video Retrieval (arxiv.org)

A Framework for Graph-Conditioned Hierarchical Shapley Attribution in Patent Valuation (arxiv.org)

pcbGPT: Automatic PCB Schematic Synthesis from Natural Language Requirements (arxiv.org)

Low-Resource Safety Failures Are Action Failures, Not Representation Failures (arxiv.org)

Implicit Geographic Inference in LLM Medical Triage: Language-Driven Disparities in Emergency Recommendations (arxiv.org)

DiscourseFlip: An Oblique Discourse-Level Opinion Manipulation Attack against Black-box Retrieval-Augmented Generation (arxiv.org)

TECCI: Tricky Edits of Collected and Curated Images (arxiv.org)

Towards Lightweight Reliability: Using Soft Prompts for Hallucination Mitigation in Large Language Models (arxiv.org)

Single-Channel Tissue Segmentation via Cross-Modal Distillation from Foundation Models (arxiv.org)

Machine Learning Surrogate Modeling for Homogenization of Hyperelastic Materials with Boolean Microstructures (arxiv.org)