mlsanityData to 14 Oct 2026

Unsupervised Confidence Calibration for Reasoning LLMs from a Single Generation

arXiv 2604.19444 ↗First posted 21 Apr 2026cs.LGcited by 3

On the map: LLM reasoning → Test-time compute and efficient thinking

Abstract (arXiv)

Reasoning language models can solve increasingly complex tasks, but struggle to produce the calibrated confidence estimates necessary for reliable deployment. Existing calibration methods usually depend on labels or repeated sampling at inference time, making them impractical in many settings. We introduce a method for unsupervised confidence calibration of reasoning LLMs when only a single generation is available at inference time. Our approach uses offline sampling on unlabeled data to derive a self-consistency-based proxy target, then distills this signal into a lightweight deployment-time confidence predictor. In a broad evaluation across 5 math and question-answering tasks using 9 reasoning models, our method substantially outperforms baselines, including under distribution shift, and improves downstream performance in selective prediction and simulated downstream decision-making.

Key facts. Cited by 3 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 44 of its 53 references that are in the map. First posted on arXiv 21 Apr 2026.

Builds on

44 of 53 references.

Closest ancestors · 2

Same direction, newest first

  1. 2025Reasoning about Uncertainty: Do Reasoning Models Know When They Don't Know?22 Jun
  2. 2025Reasoning Models Better Express Their Confidence20 May

Other references · 42

  1. 2021Training Verifiers to Solve Math Word Problemscited by 10,109
  2. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,944
  3. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,708
  4. 2017On Calibration of Modern Neural Networkscited by 4,960
  5. 2015Domain-Adversarial Training of Neural Networkscited by 3,044
  6. 2017TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehensioncited by 2,955
  7. 2019Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shiftcited by 1,287
  8. 2023Can LLMs Express Their Uncertainty? An Empirical Evaluation of Confidence Elicitation in LLMscited by 974
  9. 2017Selective Classification for Deep Neural Networkscited by 852
  10. 2023Semantic Uncertainty: Linguistic Invariances for Uncertainty Estimation in Natural Language Generationcited by 819
  11. 2023Just Ask for Calibration: Strategies for Eliciting Calibrated Confidence Scores from Language Models Fine-Tuned with Human Feedbackcited by 812
  12. 2017Crowdsourcing Multiple Choice Science Questionscited by 765
30 more
  1. 2021Revisiting the Calibration of Modern Neural Networkscited by 390
  2. 2019SelectiveNet: A Deep Neural Network with an Integrated Reject Optioncited by 350
  3. 2024INSIDE: LLMs' Internal States Retain the Power of Hallucination Detectioncited by 323
  4. 2020Calibration of Pre-trained Transformerscited by 315
  5. 2019Verified Uncertainty Calibrationcited by 309
  6. 2023Generating with Confidence: Uncertainty Quantification for Black-box Large Language Modelscited by 293
  7. 2022Prompting GPT-3 To Be Reliablecited by 260
  8. 2024Semantic Entropy Probes: Robust and Cheap Hallucination Detection in LLMscited by 259
  9. 2020Unsupervised Quality Estimation for Neural Machine Translationcited by 211
  10. 2025AIMO-2 Winning Solution: Building State-of-the-Art Mathematical Reasoning Models with OpenMathReasoning datasetcited by 174
  11. 2023Shifting Attention to Relevance: Towards the Predictive Uncertainty Quantification of Free-Form Large Language Modelscited by 150
  12. 2025AbstentionBench: Reasoning LLMs Fail on Unanswerable Questionscited by 145
  13. 2025Beyond Binary Rewards: Training LMs to Reason About Their Uncertaintycited by 117
  14. 2024Taming Overconfidence in LLMs: Reward Calibration in RLHFcited by 112
  15. 2022Uncertainty Quantification with Pre-trained Language Models: A Large-Scale Empirical Analysiscited by 104
  16. 2022Deep Unsupervised Domain Adaptation: A Review of Recent Advances and Perspectivescited by 98
  17. 2024Large language model validity via enhanced conformal prediction methodscited by 92
  18. 2023Navigating the Grey Area: How Expressions of Uncertainty and Overconfidence Affect Language Modelscited by 91
  19. 2024Calibrating Large Language Models with Sample Consistencycited by 73
  20. 2025PolyMath: Evaluating Mathematical Reasoning in Multilingual Contextscited by 56
  21. 2024Conformal Prediction for Natural Language Processing: A Surveycited by 53
  22. 2025Rewarding Doubt: A Reinforcement Learning Approach to Calibrated Confidence Expression of Large Language Modelscited by 52
  23. 2025Empowering Edge Intelligence: A Comprehensive Survey on On-Device AI Modelscited by 46
  24. 2025An Overview of Large Language Models for Statisticianscited by 24
  25. 2025Your Pre-trained LLM is Secretly an Unsupervised Confidence Calibratorcited by 20
  26. 2024To Rely or Not to Rely? Evaluating Interventions for Appropriate Reliance on Large Language Modelscited by 18
  27. 2023Prompt Risk Control: A Rigorous Framework for Responsible Deployment of Large Language Modelscited by 17
  28. 2026BaseCal: Unsupervised Confidence Calibration via Base Model Signalscited by 8
  29. 2024Improving Predictor Reliability with Selective Recalibrationcited by 3
  30. 2026Reliable and Responsible Foundation Models: A Comprehensive Surveycited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 3.

3 papers.

  1. 2026Uncertainty Quantification for LLM Agents: A Taxonomy, an Evaluation Protocol, and an Empirical Studycited by 0
  2. 2026Calibration, Not Answer Selection: Distilling Internal Confidence in Reasoning Modelscited by 0
  3. 2026Test-time Calibration Learning for Large Language Model Reasoningcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Confidence over Time: Confidence Calibration with Temporal Logic for Large Language Model Reasoning
  2. 2026Probability is Not Enough: Exploring and Counting Divergent Tokens for Reasoning Uncertainty Quantification in LLMs
  3. 2026Diagnosing Multi-step Reasoning Failures in Black-box LLMs via Stepwise Confidence Attribution
  4. 2026RL-ARC: Calibrating Large Reasoning Models via Reasoning-guided Uncertainty
  5. 2025Sample Smart, Not Hard: Correctness-First Decoding for Better Reasoning in LLMs
  6. 2025Think Just Enough: Sequence-Level Entropy as a Confidence Signal for LLM Reasoning
  7. 2025Verbalized Confidence Triggers Self-Verification: Emergent Behavior Without Explicit Reasoning Supervision
  8. 2025CER: Confidence Enhanced Reasoning in LLMs