# Of Chains and Gains: What Reinforcement Learning changes on Reasoning-based Image Quality Assessment

Authors: Yixuan Li, Yuxuan Jiang, Hanwei Zhu, Hancheng Zhu, Fan Zhang, Shiqi Wang, David Bull.

Tests how image-quality scores depend on written reasoning and its hidden states before and after reinforcement learning. Chain swaps, filler controls and activation steering separate information that can be read from a state from information that affects the score.

First public: 2026-10-23.
ArXiv initial submission: 2026-10-23.

[Original paper](https://arxiv.org/abs/2610.33203) · [PDF](https://arxiv.org/pdf/2610.33203) · [Reading-list entry](https://latent-implicit-thinking.github.io/papers/2610.33203/)
[Preprint BibTeX](https://latent-implicit-thinking.github.io/papers/2610.33203/citation.bib) · [Preprint CSL-JSON](https://latent-implicit-thinking.github.io/papers/2610.33203/citation.csl.json) · [JSON record](https://latent-implicit-thinking.github.io/papers/2610.33203/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 0.12.0; literature checked through 2026-11-02.