The Rogue Scalpel: Activation Steering Compromises LLM Safety Paper • 2509.22067 • Published Sep 26 • 27
OrtSAE: Orthogonal Sparse Autoencoders Uncover Atomic Features Paper • 2509.22033 • Published Sep 26 • 18
When Models Lie, We Learn: Multilingual Span-Level Hallucination Detection with PsiloQA Paper • 2510.04849 • Published 23 days ago • 108
Gaze into the Heart: A Multi-View Video Dataset for rPPG and Health Biomarkers Estimation Paper • 2508.17924 • Published Aug 25 • 14
3MDBench: Medical Multimodal Multi-agent Dialogue Benchmark Paper • 2504.13861 • Published Mar 26 • 2