AI News

New Evidence, Same Choice: Testing Physical Experiment Selection in Vision Language Models (arxiv.org)

A variational physics-informed graph neural network for heterogeneous solid mechanics (arxiv.org)

Building py-kvcache: A Performance Characterization of External KV Caching for vLLM with NVMe SSDs (arxiv.org)

RiLM: Parameter-Efficient Language Modeling via Geodesic Decoding (arxiv.org)

Multilingual in Name Only? Cultural and Linguistic Weaknesses of LLMs in Urdu (arxiv.org)

Weighted Empirical Risk Minimization for Machine Learning under Long-Range Dependence: Exact Pathwise Rates and Learning-Error Geometry (arxiv.org)

A Hilbert-Valued Functional Decomposition Framework for Explaining Time-Dependent Outputs (arxiv.org)

Pairit: A Platform for Live Experiments on Human-AI Collaboration (arxiv.org)

When Auditors Fabricate: Batch-Size Degradation and Confident Hallucination in LLM Detection of Planted Document Contamination (arxiv.org)

CT-SAFR: Safe and Interpretable Chain-of-Thought Reasoning for Autonomous Robots: A Multi-Layered Verification Framework for Trustworthy AI-Driven Robotic Decision Making (arxiv.org)

Looped GPT-BERT: Trading Parameters for Computation in Small Language Modeling (arxiv.org)

Which Medical Questions Deserve Rationales? Perturbation-Sensitive Selection for Robust QA (arxiv.org)

Cascading Gradient Inversion via LT-Code Inspired Peeling in Federated Learning (arxiv.org)

Hyperbolic Geometry for Open-World Object Detection in Remote Sensing Imagery (arxiv.org)

RouteBridge: Reliability-Routed Bidirectional Distillation Between Neural Radiance Fields and 3D Gaussian Splatting (arxiv.org)

Watermarks Without Verification: AI Text Watermarking After the EU AI Act (arxiv.org)

Teacher Geometry Shapes Learnability in Teacher-Student Networks (arxiv.org)

Modality-Decoupled Federated Learning for Privacy-Preserving Embodied Intelligence in 6G (arxiv.org)

Myocardial Strain Drift Correction in Deep Learning Based Ultrasound Tracking (arxiv.org)

The Vibe Shift in Software Engineering: Evaluating AI-Led Conversational Programming for Performance, Cognition, and Responsible Adoption (arxiv.org)

Arbitrary Cipher Attacks Against Large Language Models Do Not Require Fine-Tuning (arxiv.org)

A Statistical Approach to Estimating Sample Size of Machine Learning Models (arxiv.org)

Adaptive Distributed Physical-Layer Authentication and Attack Detection in 6G Non-Terrestrial Networks via Causal Meta-Learning (arxiv.org)

From Fixed Keys to Readable Schemas: Small Language Models for Vehicle Agent Function Calls (arxiv.org)

SCCM : Stream Cruise Control Method for Automated Drift Detection and Adaptation (arxiv.org)

Edu-QuRating: Multi-Dimensional Educational Data Curation with Distilled Pairwise Judgements (arxiv.org)

Reliable Near-Field Multi-User Positioning Informed by Two-Stage MUSIC (arxiv.org)

An Experimental Evaluation of Multimodal Prompt Injection Attacks on Agentic AI Frameworks (arxiv.org)

VANTAGE-Bench: Evaluating the Infrastructure AI Gap in Vision-Language Models (arxiv.org)

Auditable Emergency Triage for Maternal and Newborn Care in India (arxiv.org)

Smart Adaptive Computing Across the Continuum: LLMs in IoT-Edge-Cloud Resource Management (arxiv.org)

Improving 5G AI-RAN MCS Selection by Predicting Retransmissions (arxiv.org)

Support Discovery With Iteratively Reweighted Least Squares for Fixed-Charge Network Flow (arxiv.org)

Voice or Stereotype? Disentangling Acoustic and Content-Based Gender in Speech-to-Speech Models (arxiv.org)

DiffLUT-Net: Differentiable Training of FPGA LUT Networks with Learnable Connectivity (arxiv.org)

No Free Checker: A Survey of Verifiers for Robot Policies (arxiv.org)

In RAG We Trust? Measuring Robustness of Retrieval-Augmented Generation Under Document Poisoning (arxiv.org)

Talking to Itself While Coding: What Makes Comments Help Code Generation? (arxiv.org)

Distribution-Consistent Inference for Dynamic Sparse Mixture-of-Experts (arxiv.org)

Scaling Post-Training Ternarisation to Qwen3-8B Capability Retention, Reproduction, Lossless Packing, and Packed Execution (arxiv.org)

When Synthetic Data Hurts: On Catastrophic Forgetting in Skill Retrieval for LLM Agents (arxiv.org)

Compute-Bounded Security Assurance - Coverage, Verification, and Response under Resource Constraints (arxiv.org)

Scores Alone Do Not Prove Discovery: The Discovery Certification Protocol for Auditing AI Research Agents (arxiv.org)

Geometry Conditioning in an Embodied SLM: Training Controls and Robustness Diagnostics in a 0.8B Hybrid Model (arxiv.org)

AgentHijack: Visual Patch Attacks on Multimodal Computer-Use Agents (arxiv.org)

Empirical Evaluation of Data Poisoning Attacks in Supervised Learning (arxiv.org)

M3-Former: Multimodal Transformer with Mixture-of-Experts for Long-Term Vessel Trajectory Prediction (arxiv.org)

Bio-inspired Learning and Decision-Making with Probabilistic In-Memory Computing Hardware: Part 1 (arxiv.org)

Autonomous-Flow-Based Generation (arxiv.org)

Learning-Based Surrogate Method for Stochastic Optimization under Decision-Dependent Uncertainty with Adaptive Random Designs (arxiv.org)