Training Continuous Chain of Thought Models: A Tale of Two Regimes
Introduces direct latent supervision through averaged trace embeddings and finds that both direct and indirect continuous-CoT training degrade sharply on tasks with substantially longer reasoning traces.
First public:
Cite this paper
ArXiv preprint
arXiv:2607.16972 · Initial submission: 2026-07-18
A published-version citation has not been verified for this entry. This does not mean the paper is unpublished.
Choose the version you used. The arXiv link may serve a revision; check the original paper for the exact result, assumptions and limitations you cite.
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.
Analysis and evaluation — Measure how well these methods work, investigate what their internal computation does, and identify where they fail. This includes experiments, studies of model internals, and mathematical results.
Read or reuse this entry
Plain-text Markdown · JSON record · Find in the full list
Our summary and categories are a reading aid, not the authors’ abstract. Use the original paper for research claims. Suggest a correction.