# ReasonFlow: Flow Matching that Thinks Step by Step

Authors: Bing Yan, Ge Sun, Juan J. de Pablo, Stefano Martiniani, Kyunghyun Cho.

Trains a flow model to follow continuous states derived from prefixes of worked solutions. At inference, it integrates a trajectory from noise using the question and decodes the final state into an answer.

First public: 2026-10-24.

[Original paper](https://aclanthology.org/2026.findings-emnlp.2268/) · [Reading-list entry](https://latent-implicit-thinking.github.io/papers/yan2026reasonflow/)
[BibTeX](https://latent-implicit-thinking.github.io/papers/yan2026reasonflow/citation.bib) · [CSL-JSON](https://latent-implicit-thinking.github.io/papers/yan2026reasonflow/citation.csl.json) · [JSON record](https://latent-implicit-thinking.github.io/papers/yan2026reasonflow/record.json)

Date note: ACL Anthology release date. [Source](https://github.com/acl-org/acl-anthology/commit/635539a0ab6bdce67dae20affa31725587cf3c0c).

## Published version

ReasonFlow: Flow Matching that Thinks Step by Step. Bing Yan, Ge Sun, Juan J. de Pablo, Stefano Martiniani, Kyunghyun Cho. Findings of the Association for Computational Linguistics: EMNLP 2026, 2026.
[Publication](https://aclanthology.org/2026.findings-emnlp.2268/) · [BibTeX](https://latent-implicit-thinking.github.io/papers/yan2026reasonflow/publication.bib) · [CSL-JSON](https://latent-implicit-thinking.github.io/papers/yan2026reasonflow/publication.csl.json)
[Publication record](https://aclanthology.org/2026.findings-emnlp.2268/).

## Approaches

- [Continuous-state reasoning](https://latent-implicit-thinking.github.io/topics/continuous-states/): 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.
- [Alternative generation methods](https://latent-implicit-thinking.github.io/topics/alternative-reasoning-schedules/): Change how intermediate steps are generated, revised, or searched—for example, by refining several steps in parallel. Some produce written chains of thought.

Reading-list version 0.15.2; literature checked through 2026-11-03.