Incorporating Explicit and Implicit Chain-of-Thought as Hybrid Reasoning Paradigm with Guided Latent Space Modeling
Predicts separate continuous guide states for reasoning and answering from the question, trained against pooled embeddings of the corresponding text. Inserts these states around an explicit chain of thought during decoding.
First public: · Added to this list: November 9, 2026
ACL Anthology release date. Date source.
Cite this paper
Published version
Incorporating Explicit and Implicit Chain-of-Thought as Hybrid Reasoning Paradigm with Guided Latent Space Modeling
Guangyu Li, Zhekai Ge, Zhuo Chen, Zhen Zhang, Xinyu Wang, Yong Jiang, Kewei Tu
Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, 2026, pp. 32329-32347.
Explore this approach
Continuous-state reasoning — Use vectors—lists of numbers—to carry intermediate reasoning from one step to the next, instead of selecting a text token at every step. Some methods use the model's hidden states; others combine several candidate token representations.
Internalized reasoning — Learn from worked solutions so that some or all of the reasoning can later happen without writing those steps. Training may use a separate teacher model or progressively remove written steps from the model being trained.