# Read Your Own Mind: Reasoning Helps Surface Self-Confidence Signals in LLMs

arXiv: https://arxiv.org/abs/2505.23845 · HTML: https://mlsanity.com/paper/2505.23845/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning.
- Builds on 23 of its 28 references that are in the map.
- First posted on arXiv 28 May 2025.

## Placed in

- LLM reasoning / [Test-time compute and efficient thinking](https://mlsanity.com/lr/test-time-compute-and-efficient-thinking/)

## Builds on (most cited first, 23 of 23)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehension](https://mlsanity.com/paper/1705.03551/) (2017)
- [Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters](https://mlsanity.com/paper/2408.03314/) (2024)
- [A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questions](https://mlsanity.com/paper/2311.05232/) (2023)
- [Can LLMs Express Their Uncertainty? An Empirical Evaluation of Confidence Elicitation in LLMs](https://mlsanity.com/paper/2306.13063/) (2023)
- [Just Ask for Calibration: Strategies for Eliciting Calibrated Confidence Scores from Language Models Fine-Tuned with Human Feedback](https://mlsanity.com/paper/2305.14975/) (2023)
- [Measuring short-form factuality in large language models](https://mlsanity.com/paper/2411.04368/) (2024)
- [Embers of Autoregression: Understanding Large Language Models Through the Problem They are Trained to Solve](https://mlsanity.com/paper/2309.13638/) (2023)
- [Sources of Hallucination by Large Language Models on Inference Tasks](https://mlsanity.com/paper/2305.14552/) (2023)
- [On Verbalized Confidence Scores for LLMs](https://mlsanity.com/paper/2412.14737/) (2024)
- [Mutual Information Alleviates Hallucinations in Abstractive Summarization](https://mlsanity.com/paper/2210.13210/) (2022)
- [Linguistic Calibration of Long-Form Generations](https://mlsanity.com/paper/2404.00474/) (2024)
- [Calibrating the Confidence of Large Language Models by Eliciting Fidelity](https://mlsanity.com/paper/2404.02655/) (2024)
- [Faithfulness-Aware Decoding Strategies for Abstractive Summarization](https://mlsanity.com/paper/2303.03278/) (2023)
- [Fact-and-Reflection (FaR) Improves Confidence Calibration of Large Language Models](https://mlsanity.com/paper/2402.17124/) (2024)
- [Are Large Language Models More Honest in Their Probabilistic or Verbalized Confidence?](https://mlsanity.com/paper/2408.09773/) (2024)
- [Confidence in the Reasoning of Large Language Models](https://mlsanity.com/paper/2412.15296/) (2024)
- [Is That Your Final Answer? Test-Time Scaling Improves Selective Question Answering](https://mlsanity.com/paper/2502.13962/) (2025)
- [Beyond the Last Answer: Your Reasoning Trace Uncovers More than You Think](https://mlsanity.com/paper/2504.20708/) (2025)
- [Calibration of Machine Reading Systems at Scale](https://mlsanity.com/paper/2203.10623/) (2022)

## Built on by (most cited first, 10 of 10)

- [BrokenMath: A Benchmark for Sycophancy in Theorem Proving with LLMs](https://mlsanity.com/paper/2510.04721/) (2025)
- [Anthropomimetic Uncertainty: What Verbalized Uncertainty in Language Models is Missing](https://mlsanity.com/paper/2507.10587/) (2025)
- [Hallucinations Undermine Trust; Metacognition is a Way Forward](https://mlsanity.com/paper/2605.01428/) (2026)
- [SELFDOUBT: Uncertainty Quantification for Reasoning LLMs via the Hedge-to-Verify Ratio](https://mlsanity.com/paper/2604.06389/) (2026)
- [How Uncertainty Estimation Scales with Sampling in Reasoning Models](https://mlsanity.com/paper/2603.19118/) (2026)
- [Future Confidence Distillation in Large Language Models](https://mlsanity.com/paper/2607.07626/) (2026)
- [Enhancing Confidence Estimation in Telco LLMs via Twin-Pass CoT-Ensembling](https://mlsanity.com/paper/2604.13271/) (2026)
- [Operadic consistency: a label-free signal for compositional reasoning failures in LLMs](https://mlsanity.com/paper/2606.13649/) (2026)
- [Auto-Fill: Learning to Predict Missing Values Accurately with Specialist Language Models](https://mlsanity.com/paper/2607.19847/) (2026)
- [Different Facets of Verbalised Overconfidence: an Interpretability Study](https://mlsanity.com/paper/2608.18106/) (2026)

## Abstract (arXiv)

We study the source of uncertainty in DeepSeek R1-32B by analyzing its self-reported verbal confidence on question answering (QA) tasks. In the default answer-then-confidence setting, the model is regularly over-confident, whereas semantic entropy - obtained by sampling many responses - remains reliable. We hypothesize that this is because of semantic entropy's larger test-time compute, which lets us explore the model's predictive distribution. We show that granting DeepSeek the budget to explore its distribution by forcing a long chain-of-thought before the final answer greatly improves its verbal score effectiveness, even on simple fact-retrieval questions that normally require no reasoning. Furthermore, a separate reader model that sees only the chain can reconstruct very similar confidences, indicating the verbal score might be merely a statistic of the alternatives surfaced during reasoning. Our analysis concludes that reliable uncertainty estimation requires explicit exploration of the generative space, and self-reported confidence is trustworthy only after such exploration.