Tong Bai
1 indexed paper
Recent (6 mo)
1With code
0Influential cites
0Benchmarked
0Publications per year
126
Top categories
NLP×1ML×1
Frequent co-authors
Research Timeline
2026
Unlocking the Black Box of Latent Reasoning: An Interpretability-Guided Approach to Intervention
This paper introduces interpretability-guided, training-free interventions that systematically improve the accuracy and controllability of latent reasoning in LLMs by leveraging structural and causal insights into continuous hidden states.
Highlighted terms show continued research focus across papers
Papers
cs.CLcs.LGRecentMay 31, 2026
Unlocking the Black Box of Latent Reasoning: An Interpretability-Guided Approach to Intervention
Shuochen Chang, Tong Bai, Xiaofeng Zhang, Qianli Ma +4 more
This paper introduces interpretability-guided, training-free interventions that systematically improve the accuracy and controllability of latent reasoning in LLMs by leveraging structural and causal…
View →