# Do Reasoning Representations Help Humans Evaluate LLM Outputs?

arXiv: https://arxiv.org/abs/2609.09038 · HTML: https://mlsanity.com/paper/2609.09038/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning.
- Builds on 10 of its 53 references that are in the map.
- First posted on arXiv 8 Sep 2026.

## Placed in

- LLM reasoning / [Reasoning benchmarks and evaluation](https://mlsanity.com/lr/reasoning-benchmarks-and-evaluation/)

## Builds on (most cited first, 10 of 10)

- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Teaching Large Language Models to Self-Debug](https://mlsanity.com/paper/2304.05128/) (2023)
- [Show Your Work: Scratchpads for Intermediate Computation with Language Models](https://mlsanity.com/paper/2112.00114/) (2021)
- [Training Large Language Models to Reason in a Continuous Latent Space](https://mlsanity.com/paper/2412.06769/) (2024)
- [Measuring Faithfulness in Chain-of-Thought Reasoning](https://mlsanity.com/paper/2307.13702/) (2023)
- [Thought Anchors: Which LLM Reasoning Steps Matter?](https://mlsanity.com/paper/2506.19143/) (2025)
- [Cumulative Reasoning with Large Language Models](https://mlsanity.com/paper/2308.04371/) (2023)
- [Demystifying Chains, Trees, and Graphs of Thoughts](https://mlsanity.com/paper/2401.14295/) (2024)
- [Evaluating LLMs at Detecting Errors in LLM Responses](https://mlsanity.com/paper/2404.03602/) (2024)
- [Improving Human Verification of LLM Reasoning through Interactive Explanation Interfaces](https://mlsanity.com/paper/2510.22922/) (2025)

## Abstract (arXiv)

Reasoning representations are increasingly used as explanations for large language model outputs. Yet they are typically evaluated with model-centric criteria, such as answer accuracy and faithfulness, leaving it unclear whether they help people evaluate model responses. In this work, we study reasoning representations as human-facing interfaces rather than proxies for model reasoning ability. We conduct a controlled human study of six reasoning formats across tasks of varying complexity, supported by a web-based framework that randomizes task domains, problem instances, and representation order. The study collects fine-grained judgments of structural understanding, error detection and localization, and trust calibration. Our study shows a mismatch between perceived preference and support for human evaluation. Participants prefer planning- and decomposition-based representations, but simpler chain-of-thought traces better support verification, trust, and interpretability. Preferred representations also introduce calibration risks, with more false alarms on correct traces and high trust despite low willingness to verify.