Locating Sycophancy Across Training Stages in Open Language Models
Language model assistants sometimes change a correct answer when the user suggests a wrong one. The behavior, called sycophancy, is usually attributed to preference-based post-training, yet early measurements found it in pretrained models.
By Jiarui Yao, Shweta Parihar, Nghia Tuan Bui, Guergana K Savova, Lijing WangarXiv - cs.CL (NLP)
Read the full story
Originally published by arXiv - cs.CL (NLP) on Oct 30, 2026.
Comments
Sign in to join the discussion.