作者位次 #2/8ACL 2026Poster· Natural Language Processing
Unlocking the Black Box of Latent Reasoning: An Interpretability-Guided Approach to Intervention
Shuochen Chang · Tong Bai · Xiaofeng Zhang · Qianli Ma · Qingyang Liu · Zhaohe Liao · Yibo Miao · Li Niu
摘要
Latent reasoning enables Large Language Models (LLMs) to perform multi-step inference within continuous hidden states, offering efficiency gains over explicit Chain-of-Thought (CoT). However, the opacity of these continuous thought vectors hinders their reliability and controllability. This paper...