Zonghao Ying

PhD student, Beijing University of Aeronautics and Astronautics

1 paper at NeurIPS 2025

Homepage· OpenReview· Semantic Scholar· Google Scholar

Poster Session 4

1 paper

Thursday, December 4, 2025 · 4:30 PM → 7:30 PM

Exhibit Hall C,D,E

Detoxifying Large Language Models via Autoregressive Reward Guided Representation Editing

#1109 · Yisong Xiao, Aishan Liu, Siyuan Liang, Zonghao Ying, Xianglong Liu, Dacheng Tao

We propose a test-time detoxification framework that models toxicity transitions within the latent representation space to enable stable and precise representation editing guidance.