Legibility is Not Interpretability: Comparing Judged and Actual Importance in Chain-Of-Thought Reasoning

• URL: http://arxiv.org/abs/2609.04194v1
• Date: Thu, 03 Sep 2026 17:59:08 GMT
• Title: Legibility is Not Interpretability: Comparing Judged and Actual Importance in Chain-Of-Thought Reasoning
• Abstract summary: Reasoning traces from chain-of-thought models appear to offer a legible window into how a model arrives at its answer.<n>We operationalize the importance of a reasoning step as its advantage.<n>We find that sufficiently capable LLMs can outperform a prevalence baseline but fall well short of a noise ceiling.
• Score: 48.96460246231164
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Reasoning traces from chain-of-thought models appear to offer a legible window into how a model arrives at its answer. A growing body of work treats them as such, using LLM judges to diagnose errors, evaluate faithfulness, and provide step-level supervision via process reward models and generative critics. These practices rely on the text of a reasoning step carrying information about its functional role. But does the text actually encode information about which reasoning steps matter? We operationalize the importance of a reasoning step as its advantage: the change in expected reward, e.g., producing the correct final answer, from including that step, estimated via Monte Carlo rollouts. Basing ground truth on these estimates, we evaluate whether LLM judges can identify high-advantage steps and find that sufficiently capable LLMs can outperform a prevalence baseline but fall well short of a noise ceiling. Fine-tuning a model as a step-level critic yields strong improvement for incorrect responses but remains distant from ceiling for correct responses, suggesting that step importance is only partially recoverable from the text of the reasoning trace. Our findings contribute to a growing body of chain-of-thought faithfulness work that cautions against treating the legibility of reasoning traces as interpretability, especially with implications for process reward modeling.

Related papers

• Balancing Faithfulness and Performance in Reasoning via Multi-Listener Soft Execution [79.98699884805636]
  Reasoning Execution by Multiple Listeners (REMUL) is a multi-party reinforcement learning approach.<n>REMUL builds on the hypothesis that reasoning traces which other parties can follow will be more faithful.<n>Speakers are rewarded for producing reasoning that is clear to listeners.
  arXiv  Detail & Related papers  (2026-02-18T02:55:55Z)
• Stop Rewarding Hallucinated Steps: Faithfulness-Aware Step-Level Reinforcement Learning for Small Reasoning Models [59.6715047267181]
  Small reasoning models (SRMs) are prone to hallucinations, especially in intermediate reasoning steps.<n>Existing mitigation methods based on online reinforcement learning rely on outcome-based rewards or coarse-grained chain-of-thought evaluation.<n>We propose Faithfulness-Aware Step-Level Reinforcement Learning (FaithRL), introducing step-level supervision via explicit faithfulness rewards from a process reward model.
  arXiv  Detail & Related papers  (2026-02-05T17:15:12Z)
• Gaming the Judge: Unfaithful Chain-of-Thought Can Undermine Agent Evaluation [76.5533899503582]
  Large language models (LLMs) are increasingly used as judges to evaluate agent performance.<n>We show this paradigm implicitly assumes that the agent's chain-of-thought (CoT) reasoning faithfully reflects both its internal reasoning and the underlying environment state.<n>We demonstrate that manipulated reasoning alone can inflate false positive rates of state-of-the-art VLM judges by up to 90% across 800 trajectories spanning diverse web tasks.
  arXiv  Detail & Related papers  (2026-01-21T06:07:43Z)
• Journey Before Destination: On the importance of Visual Faithfulness in Slow Thinking [11.763473690046721]
  Reasoning-augmented vision language models generate explicit chains of thought that promise greater capability and transparency.<n>Models may reach correct answers via visually unfaithful intermediate steps, or reason faithfully yet fail on the final prediction.<n>We introduce the visual faithfulness of reasoning chains as a distinct evaluation dimension, focusing on whether the perception steps of a reasoning chain are grounded in the image.
  arXiv  Detail & Related papers  (2025-12-13T07:04:42Z)
• What Defines Good Reasoning in LLMs? Dissecting Reasoning Steps with Multi-Aspect Evaluation [67.47463575774388]
  We decompose reasoning quality into two dimensions: relevance and coherence.<n>To measure these aspects reliably, we introduce causal stepwise evaluation (CaSE)<n>We show that curating training data with CaSE-evaluated relevance and coherence directly improves final task performance.
  arXiv  Detail & Related papers  (2025-10-23T14:30:37Z)
• Making Reasoning Matter: Measuring and Improving Faithfulness of Chain-of-Thought Reasoning [38.60086807496399]
  Large language models (LLMs) have been shown to perform better when asked to reason step-by-step before answering a question. It is unclear to what degree the model's final answer is faithful to the stated reasoning steps. We introduce FRODO, a framework to tailor small-sized LMs to generate correct reasoning steps and robustly reason over these steps.
  arXiv  Detail & Related papers  (2024-02-21T17:23:59Z)
• A Closer Look at the Self-Verification Abilities of Large Language Models in Logical Reasoning [73.77088902676306]
  We take a closer look at the self-verification abilities of large language models (LLMs) in the context of logical reasoning. Our main findings suggest that existing LLMs could struggle to identify fallacious reasoning steps accurately and may fall short of guaranteeing the validity of self-verification methods.
  arXiv  Detail & Related papers  (2023-11-14T07:13:10Z)
• Question Decomposition Improves the Faithfulness of Model-Generated Reasoning [23.34325378824462]
  Large language models (LLMs) are difficult to verify the correctness and safety of their behavior. One approach is to prompt LLMs to externalize their reasoning, by having them generate step-by-step reasoning as they answer a question. This approach relies on the stated reasoning faithfully reflecting the model's actual reasoning, which is not always the case. Decomposition-based methods achieve strong performance on question-answering tasks, sometimes approaching that of CoT.
  arXiv  Detail & Related papers  (2023-07-17T00:54:10Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.