mlsanityData to 14 Oct 2026

Analyzing Uncertainty of LLM-as-a-Judge: Interval Evaluations with Conformal Prediction

arXiv 2509.18658 ↗First posted 23 Sep 2025cs.CLcited by 27

Not on the map. Why?

Abstract (arXiv)

LLM-as-a-judge has become a promising paradigm for using large language models (LLMs) to evaluate natural language generation (NLG), but the uncertainty of its evaluation remains underexplored. This lack of reliability may limit its deployment in many applications. This work presents the first framework to analyze the uncertainty by offering a prediction interval of LLM-based scoring via conformal prediction. Conformal prediction constructs continuous prediction intervals from a single evaluation run, and we design an ordinal boundary adjustment for discrete rating tasks. We also suggest a midpoint-based score within the interval as a low-bias alternative to raw model score and weighted average. We perform extensive experiments and analysis, which show that conformal prediction can provide valid prediction interval with coverage guarantees. We also explore the usefulness of interval midpoint and judge reprompting for better judgment.

Key facts. Cited by 27 papers in the map (data through 14 Oct 2026). Builds on 50 of its 74 references that are in the map. First posted on arXiv 23 Sep 2025.

Builds on

50 of 74 references.

References · 50

  1. 2021Training Verifiers to Solve Math Word Problemscited by 10,109
  2. 2019BERTScore: Evaluating Text Generation with BERTcited by 5,512
  3. 2023G-Eval: NLG Evaluation using GPT-4 with Better Human Alignmentcited by 2,276
  4. 2024A Survey on LLM-as-a-Judgecited by 1,423
  5. 2023Can LLMs Express Their Uncertainty? An Empirical Evaluation of Confidence Elicitation in LLMscited by 974
  6. 2019DROP: A Reading Comprehension Benchmark Requiring Discrete Reasoning Over Paragraphscited by 903
  7. 2023Just Ask for Calibration: Strategies for Eliciting Calibrated Confidence Scores from Language Models Fine-Tuned with Human Feedbackcited by 812
  8. 2021A Gentle Introduction to Conformal Prediction and Distribution-Free Uncertainty Quantificationcited by 792
  9. 2020SummEval: Re-evaluating Summarization Evaluationcited by 674
  10. 2019Conformalized Quantile Regressioncited by 629
  11. 2018e-SNLI: Natural Language Inference with Natural Language Explanationscited by 468
  12. 2020Classification with Valid and Adaptive Coveragecited by 396
38 more
  1. 2020Uncertainty Sets for Image Classifiers using Conformal Predictioncited by 366
  2. 2021Learn then Test: Calibrating Predictive Algorithms to Achieve Risk Controlcited by 271
  3. 2022ROSCOE: A Suite of Metrics for Scoring Step-by-Step Reasoningcited by 177
  4. 2023Conformal Language Modelingcited by 170
  5. 2023Conformal Prediction with Large Language Models for Multi-Choice Question Answeringcited by 139
  6. 2024A Note on Shumailov et al. (2024): `AI Models Collapse When Trained on Recursively Generated Data'cited by 115
  7. 2021Localized Conformal Prediction: A Generalized Inference Framework for Conformal Predictioncited by 114
  8. 2024SaySelf: Teaching LLMs to Express Confidence with Self-Reflective Rationalescited by 105
  9. 2024Trust or Escalate: LLM Judges with Provable Guarantees for Human Agreementcited by 105
  10. 2024Can Large Language Models Faithfully Express Their Intrinsic Uncertainty in Words?cited by 87
  11. 2021Conformal Prediction using Conditional Histogramscited by 77
  12. 2019A comparison of some conformal quantile regression methodscited by 76
  13. 2024API Is Enough: Conformal Prediction for Large Language Models Without Logit-Accesscited by 69
  14. 2024ConU: Conformal Uncertainty in Large Language Models with Correctness Coverage Guaranteescited by 65
  15. 2024Can You Trust LLM Judgments? Reliability of LLM-as-a-Judgecited by 56
  16. 2024Conformal Prediction for Natural Language Processing: A Surveycited by 53
  17. 2024Calibrating the Confidence of Large Language Models by Eliciting Fidelitycited by 47
  18. 2025RocketEval: Efficient Automated LLM Evaluation via Grading Checklistcited by 39
  19. 2022Improving Trustworthiness of AI Disease Severity Rating in Medical Imaging with Ordinal Conformal Prediction Setscited by 38
  20. 2025Improving LLM-as-a-Judge Inference with the Judgment Distributioncited by 37
  21. 2024CalibraEval: Calibrating Prediction Distribution to Mitigate Selection Bias in LLMs-as-Judgescited by 34
  22. 2023SocREval: Large Language Models with the Socratic Method for Reference-Free Reasoning Evaluationcited by 27
  23. 2024Fairer Preferences Elicit Improved Human-Aligned Large Language Model Judgmentscited by 25
  24. 2021Locally Valid and Discriminative Prediction Intervals for Deep Learning Modelscited by 24
  25. 2025An Empirical Analysis of Uncertainty in Large Language Model Evaluationscited by 20
  26. 2025Crowd Comparative Reasoning: Unlocking Comprehensive Evaluations for LLM-as-a-Judgecited by 19
  27. 2024Conformal Prediction via Regression-as-Classificationcited by 18
  28. 2024Black-box Uncertainty Quantification Method for LLM-as-a-Judgecited by 18
  29. 2024Prune 'n Predict: Optimizing LLM Decision-making with Conformal Predictioncited by 18
  30. 2022MAPIE: an open-source library for distribution-free uncertainty quantificationcited by 17
  31. 2025VeriFact: Verifying Facts in LLM-Generated Clinical Text with Electronic Health Recordscited by 17
  32. 2024DnA-Eval: Enhancing Large Language Model Evaluation through Decomposition and Aggregationcited by 12
  33. 2019Teaching Pretrained Models with Commonsense Reasoning: A Preliminary KB-Based Approachcited by 10
  34. 2024Conformal Generative Modeling with Improved Sample Efficiency through Sequential Greedy Filteringcited by 10
  35. 2025Expect the Unexpected: FailSafe Long Context QA for Financecited by 7
  36. 2025One Sample is Enough to Make Conformal Prediction Robustcited by 4
  37. 2025A Cost-Effective LLM-based Approach to Identify Wildlife Trafficking in Online Marketplacescited by 3
  38. 2025Aligning Black-box Language Models with Human Judgmentscited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 26.

27 papers.

Sort:

By area:

  1. 2026WebCoderBench: Benchmarking Web Application Generation with Comprehensive and Interpretable Evaluation Metricscited by 18
  2. 2026Diagnosing the Reliability of LLM-as-a-Judge via Item Response Theorycited by 16
  3. 2026VLM Judges Can Rank but Cannot Score: Task-Dependent Uncertainty in Multimodal Evaluationcited by 8
  4. 2026Adaptive Stopping for Multi-Turn LLM Reasoningcited by 7
  5. 2026Geometry-Calibrated Conformal Abstention for Language Modelscited by 5
  6. 2026URAG: A Benchmark for Uncertainty Quantification in Retrieval-Augmented Large Language Modelscited by 3
  7. 2026Beyond LLM-as-a-Judge: Deterministic Metrics for Multilingual Generative Text Evaluationcited by 3
  8. 2026Label Effects: Shared Heuristic Reliance in Trust Assessment by Humans and LLM-as-a-Judgecited by 3
  9. 2026From Actions to Understanding: Conformal Interpretability of Temporal Concepts in LLM Agentscited by 3
  10. 2025Robust Uncertainty Quantification for Self-Evolving Large Language Models via Continual Domain Pretrainingcited by 2
  11. 2026A Finite-Calibration Regime Map for LLM Judge Panelscited by 2
  12. 2026From Uncertain Judgments to Calibrated Rankings: Conformal Elo Estimation for LLM Evaluationcited by 2
15 more
  1. 2026Towards Provably Unbiased LLM Judges via Bias-Bounded Evaluationcited by 1
  2. 2026PaperJury: Due-Process Review for Bounded LaTeX Revisioncited by 1
  3. 2026Multimodal Learning on Low-Quality Data with Conformal Predictive Self-Calibrationcited by 0
  4. 2026An Interpretable and Scalable Framework for Evaluating Large Language Modelscited by 0
  5. 2026The Ghost Annotator: a Framework to Explore Human Label Variation in Content Moderation through Conformal Predictioncited by 0
  6. 2026Data-Centric Benchmarking of Exploit Generation in LLMs: Understanding the Impact of Fine-Tuningcited by 0
  7. 2026Judgment-Grounded Expansion for Peer Review Generationcited by 0
  8. 2026When Trust Meets Truth: Trust-Truth Separability in LLM-as-Judgecited by 0
  9. 2026Robust Conformal Consensus: Multi-Agent LLM-as-a-Judge Interval Evaluation with Conformal Predictioncited by 0
  10. 2026Decomposing LLM-Judge Uncertainty to Target Expert Labelscited by 0
  11. 2026Optimizing the Score, Losing Sight of the Task: Reward Hacking Across Weights, Selection, and Promptscited by 0
  12. 2026When Accuracy Gaps Fail to Certify: Auditing Cross-Domain Recalibration of LLM Judgescited by 0
  13. 2026CS-WCP: Robust Conformal Sets for LLM-Judge Traffic Shifts with Uncertain Group Proportionscited by 0
  14. 2026Range-GRPO: Policy Optimization via Pairwise Relations among Reward Intervalscited by 0
  15. 2026Reliable Self-Evolution with Imperfect Proxy Rewardscited by 0