# Interpreting Model Reasoning through Behavioral Counterfactuals

Authors: Shawn Im, James Oldfield, Adel Bibi, Sharon Li.

Holds a model response fixed while changing the prompt, then learns hidden-state edits that reproduce the change in token probabilities. Uses these edits and activation patching to show how suggestions can influence reasoning through question and filler-token states without being mentioned in the written response.

First public: 2026-10-22.
ArXiv initial submission: 2026-10-22.

[Original paper](https://arxiv.org/abs/2610.30630) · [PDF](https://arxiv.org/pdf/2610.30630) · [HTML](https://arxiv.org/html/2610.30630v1) · [Reading-list entry](https://latent-implicit-thinking.github.io/papers/2610.30630/)
[Preprint BibTeX](https://latent-implicit-thinking.github.io/papers/2610.30630/citation.bib) · [Preprint CSL-JSON](https://latent-implicit-thinking.github.io/papers/2610.30630/citation.csl.json) · [JSON record](https://latent-implicit-thinking.github.io/papers/2610.30630/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.