# On Reasoning Correctness Directions

Authors: Munachiso Samuel Nwadike, Zayd M. K. Zuhri, Rifo Genadi, Abzal Nurgazy, Ali Mekky, Alham Fikri Aji, Zangir Iklassov, Kentaro Inui.

Studies hidden-state directions that predict correctness during written arithmetic search traces. Compares linear probes, activation differences, and optimized steering vectors, finding that readable correctness signals do not yield consistent improvements when used to steer generation.

First public: 2026-10-18.
ArXiv initial submission: 2026-10-18.

[Original paper](https://arxiv.org/abs/2610.25464) · [PDF](https://arxiv.org/pdf/2610.25464) · [HTML](https://arxiv.org/html/2610.25464v1) · [Reading-list entry](https://latent-implicit-thinking.github.io/papers/2610.25464/)
[Preprint BibTeX](https://latent-implicit-thinking.github.io/papers/2610.25464/citation.bib) · [Preprint CSL-JSON](https://latent-implicit-thinking.github.io/papers/2610.25464/citation.csl.json) · [JSON record](https://latent-implicit-thinking.github.io/papers/2610.25464/record.json)

## Approaches

- [Analysis and evaluation](https://latent-implicit-thinking.github.io/topics/implicit-reasoning-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.

Reading-list version 0.11.0; literature checked through 2026-10-26.