# Energy-Based Reasoners: Your Recurrent Reasoner Should Be an Energy-Based Model

Authors: William Xingxu Chen, Alexi Gladstone, Zeyu Michael Li, Heng Ji, Yilun Du.

Refines puzzle-solving states by descending a learned energy function. Trains convergence to track answer correctness and supervises the best-matching trajectories when a problem has several valid solutions.

First public: 2026-11-06.
ArXiv initial submission: 2026-11-06.

[Original paper](https://arxiv.org/abs/2611.08977) · [PDF](https://arxiv.org/pdf/2611.08977) · [HTML](https://arxiv.org/html/2611.08977v1) · [Reading-list entry](https://latent-implicit-thinking.github.io/papers/2611.08977/)
[Preprint BibTeX](https://latent-implicit-thinking.github.io/papers/2611.08977/citation.bib) · [Preprint CSL-JSON](https://latent-implicit-thinking.github.io/papers/2611.08977/citation.csl.json) · [JSON record](https://latent-implicit-thinking.github.io/papers/2611.08977/record.json)

## Approaches

- [Recurrent reasoning](https://latent-implicit-thinking.github.io/topics/recurrent-reasoning/): Repeat parts of the model or keep intermediate information in memory while working through a problem. Some methods also learn how many passes to run.

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