一作NeurIPS 2026Poster
Steering Externalities: Benign Activation Steering Unintentionally Increases Jailbreak Risk for Large Language Models
Chen Xiong · Zhiyuan HE · Pin-Yu Chen · Ching-Yun Ko · Tsung-Yi Ho
首次获取该候选人的完整档案及 StartLux 评估共 20 积分；每位用户单独解锁，之后查看免费。主动刷新 20 积分。
实时查找电子邮箱，费用不超过20积分；未找到邮箱不扣费。
Chen Xiong · Zhiyuan HE · Pin-Yu Chen · Ching-Yun Ko · Tsung-Yi Ho
Chen Xiong · Zihao Wang · Rui Zhu · Tsung-Yi Ho · Pin-Yu Chen · Jingwei Xiong · Haixu Tang
Large Language Models (LLMs) rely on massive training datasets, often including proprietary data, which raises concerns about unauthorized usage and copyright infringement. Existing dataset inference methods typically require access to log probabilities or other internal signals, but many modern...