Picture for Zehua Cao

Zehua Cao

Faithful-Patchscopes: Understanding and Mitigating Model Bias in Hidden Representations Explanation of Large Language Models

Add code
Jan 30, 2026
Viaarxiv icon