When Models Lie, We Learn: Multilingual Span-Level Hallucination Detection with PsiloQA Paper • 2510.04849 • Published Oct 6 • 113
OrtSAE: Orthogonal Sparse Autoencoders Uncover Atomic Features Paper • 2509.22033 • Published Sep 26 • 18
The Rogue Scalpel: Activation Steering Compromises LLM Safety Paper • 2509.22067 • Published Sep 26 • 27
When Punctuation Matters: A Large-Scale Comparison of Prompt Robustness Methods for LLMs Paper • 2508.11383 • Published Aug 15 • 41
T-LoRA: Single Image Diffusion Model Customization Without Overfitting Paper • 2507.05964 • Published Jul 8 • 119
Geopolitical biases in LLMs: what are the "good" and the "bad" countries according to contemporary language models Paper • 2506.06751 • Published Jun 7 • 71