# Do Cognitively Interpretable Reasoning Traces Improve LLM Performance?

arXiv: https://arxiv.org/abs/2508.16695 · HTML: https://mlsanity.com/paper/2508.16695/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Placed in Prompting and chain-of-thought reasoning, one of 17 directions in LLM reasoning.
- Builds on 19 of its 27 references that are in the map.
- First posted on arXiv 21 Aug 2025.

## Placed in

- LLM reasoning / [Prompting and chain-of-thought reasoning](https://mlsanity.com/lr/prompting-and-chain-of-thought-reasoning/)

## Builds on (most cited first, 19 of 19)

- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [QLoRA: Efficient Finetuning of Quantized LLMs](https://mlsanity.com/paper/2305.14314/) (2023)
- [Towards A Rigorous Science of Interpretable Machine Learning](https://mlsanity.com/paper/1702.08608/) (2017)
- [A Survey Of Methods For Explaining Black Box Models](https://mlsanity.com/paper/1802.01933/) (2018)
- [Measuring Faithfulness in Chain-of-Thought Reasoning](https://mlsanity.com/paper/2307.13702/) (2023)
- [Faithful Chain-of-Thought Reasoning](https://mlsanity.com/paper/2301.13379/) (2023)
- [Teaching Small Language Models to Reason](https://mlsanity.com/paper/2212.08410/) (2022)
- [Chain-of-Thought Reasoning In The Wild Is Not Always Faithful](https://mlsanity.com/paper/2503.08679/) (2025)
- [Distilling Reasoning Capabilities into Smaller Language Models](https://mlsanity.com/paper/2212.00193/) (2022)
- [Making Reasoning Matter: Measuring and Improving Faithfulness of Chain-of-Thought Reasoning](https://mlsanity.com/paper/2402.13950/) (2024)
- [Beyond Answers: Transferring Reasoning Capabilities to Smaller LLMs Using Multi-Teacher Knowledge Distillation](https://mlsanity.com/paper/2402.04616/) (2024)
- [Measuring Chain of Thought Faithfulness by Unlearning Reasoning Steps](https://mlsanity.com/paper/2502.14829/) (2025)
- [On the Hardness of Faithful Chain-of-Thought Reasoning in Large Language Models](https://mlsanity.com/paper/2406.10625/) (2024)
- [Beyond Semantics: The Unreasonable Effectiveness of Reasonless Intermediate Tokens](https://mlsanity.com/paper/2505.13775/) (2025)
- [Living in the Moment: Can Large Language Models Grasp Co-Temporal Reasoning?](https://mlsanity.com/paper/2406.09072/) (2024)
- [Towards Better Chain-of-Thought: A Reflection on Effectiveness and Faithfulness](https://mlsanity.com/paper/2405.18915/) (2024)
- [Interpretable Traces, Unexpected Outcomes: Investigating the Disconnect in Trace-Based Knowledge Distillation](https://mlsanity.com/paper/2505.13792/) (2025)
- [Predictability and Comprehensibility in Post-Hoc XAI Methods: A User-Centered Analysis](https://mlsanity.com/paper/2309.11987/) (2023)

## Built on by (most cited first, 11 of 11)

- [From Fluent to Verifiable: Claim-Level Auditability for Deep Research Agents](https://mlsanity.com/paper/2602.13855/) (2026)
- [Local Coherence or Global Validity? Investigating RLVR Traces in Math Domains](https://mlsanity.com/paper/2510.18176/) (2025)
- [Do explanations generalize across large reasoning models?](https://mlsanity.com/paper/2601.11517/) (2026)
- [Measuring Weak-to-Strong Legibility of Reasoning Models](https://mlsanity.com/paper/2603.20508/) (2026)
- [MentorCollab: Large-to-Small Inference-Time Mentorship for Concise Reasoning in Language Models](https://mlsanity.com/paper/2602.05307/) (2026)
- [Measuring and Mitigating Post-hoc Rationalization in Reverse Chain-of-Thought Generation](https://mlsanity.com/paper/2602.14469/) (2026)
- [The Potential of CoT for Reasoning: A Closer Look at Trace Dynamics](https://mlsanity.com/paper/2602.14903/) (2026)
- [STEMVerse: A Dual-Axis Diagnostic Framework for STEM Reasoning in Large Language Models](https://mlsanity.com/paper/2602.02497/) (2026)
- [Revisiting the UID Hypothesis in LLM Reasoning Traces](https://mlsanity.com/paper/2510.13850/) (2025)
- [Answering the Wrong Question: Reasoning Trace Inversion for Abstention in LLMs](https://mlsanity.com/paper/2604.02230/) (2026)
- [R2IF: Aligning Reasoning with Decisions via Composite Rewards for Interpretable LLM Function Calling](https://mlsanity.com/paper/2604.20316/) (2026)

## Abstract (arXiv)

Recent progress in reasoning-oriented Large Language Models (LLMs) has been driven by introducing Chain-of-Thought (CoT) traces, where models generate intermediate reasoning traces before producing an answer. These traces, as in DeepSeek R1, are not only used to guide inference but also serve as supervision signals for distillation into smaller models. A common but often implicit assumption is that CoT traces should be semantically meaningful and interpretable to the end user. While recent research questions the need for semantic nature of these traces, in this paper, we ask: ``textitMust CoT reasoning traces be interpretable to enhance LLM task performance?" We investigate this question in the Open Book Question-Answering domain by supervised fine-tuning LLaMA and Qwen models on four types of reasoning traces: (1) DeepSeek R1 traces, (2) LLM-generated summaries of R1 traces, (3) LLM-generated post-hoc explanations of R1 traces, and (4) algorithmically generated verifiably correct traces. To quantify the trade-off between interpretability and performance, we further conduct a human-subject study with 100 participants rating the interpretability of each trace type. Our results reveal a striking mismatch: while fine-tuning on R1 traces yields the strongest performance, participants judged these traces to be the least interpretable. These findings suggest that it is useful to decouple intermediate tokens from end user interpretability.