mlsanityData to 14 Oct 2026

Beyond Surface Statistics: Robust Conformal Prediction for LLMs via Internal Representations

arXiv 2604.16217 ↗First posted 17 Apr 2026cs.CL cs.AIcited by 1

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

Abstract (arXiv)

Large language models are increasingly deployed in settings where reliability matters, yet output-level uncertainty signals such as token probabilities, entropy, and self-consistency can become brittle under calibration--deployment mismatch. Conformal prediction provides finite-sample validity under exchangeability, but its practical usefulness depends on the quality of the nonconformity score. We propose a conformal framework for LLM question answering that uses internal representations rather than output-facing statistics: specifically, we introduce Layer-Wise Information (LI) scores, which measure how conditioning on the input reshapes predictive entropy across model depth, and use them as nonconformity scores within a standard split conformal pipeline. Across closed-ended and open-domain QA benchmarks, with the clearest gains under cross-domain shift, our method achieves a better validity--efficiency trade-off than strong text-level baselines while maintaining competitive in-domain reliability at the same nominal risk level. These results suggest that internal representations can provide more informative conformal scores when surface-level uncertainty is unstable under distribution shift.

Key facts. Cited by 1 paper 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 30 of its 34 references that are in the map. First posted on arXiv 17 Apr 2026.

Builds on

30 of 34 references.

Closest ancestors · 1

Same direction, newest first

  1. 2025Representation Consistency for Accurate and Coherent LLM Answer Aggregation18 Jun

Other references · 29

  1. 2017TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehensioncited by 2,955
  2. 2024MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmarkcited by 1,848
  3. 2024Qwen2.5-Math Technical Report: Toward Mathematical Expert Model via Self-Improvementcited by 1,090
  4. 2023Semantic Uncertainty: Linguistic Invariances for Uncertainty Estimation in Natural Language Generationcited by 819
  5. 2021A Gentle Introduction to Conformal Prediction and Distribution-Free Uncertainty Quantificationcited by 792
  6. 2022Discovering Latent Knowledge in Language Models Without Supervisioncited by 767
  7. 2018CoQA: A Conversational Question Answering Challengecited by 683
  8. 2023The Internal State of an LLM Knows When It's Lyingcited by 659
  9. 2022MedMCQA : A Large-scale Multi-Subject Multi-Choice Dataset for Medical domain Question Answeringcited by 543
  10. 2019Conformal Prediction Under Covariate Shiftcited by 519
  11. 2021Adaptive Conformal Inference Under Distribution Shiftcited by 375
  12. 2024INSIDE: LLMs' Internal States Retain the Power of Hallucination Detectioncited by 323
17 more
  1. 2023Generating with Confidence: Uncertainty Quantification for Black-box Large Language Modelscited by 293
  2. 2021Understanding Dataset Difficulty with V-Usable Informationcited by 291
  3. 2020A Theory of Usable Information Under Computational Constraintscited by 162
  4. 2023Conformal Prediction With Conditional Guaranteescited by 152
  5. 2023Conformal Prediction with Large Language Models for Multi-Choice Question Answeringcited by 139
  6. 2024Language Models with Conformal Factuality Guaranteescited by 131
  7. 2024Theoretical Foundations of Conformal Predictioncited by 101
  8. 2024Large language model validity via enhanced conformal prediction methodscited by 92
  9. 2024API Is Enough: Conformal Prediction for Large Language Models Without Logit-Accesscited by 69
  10. 2024ConU: Conformal Uncertainty in Large Language Models with Correctness Coverage Guaranteescited by 65
  11. 2024Probabilistic Conformal Prediction with Approximate Conditional Validitycited by 23
  12. 2025SConU: Selective Conformal Uncertainty in Large Language Modelscited by 23
  13. 2024An Information Theoretic Perspective on Conformal Predictioncited by 21
  14. 2025COIN: Uncertainty-Guarding Selective Question Answering for Foundation Models with Provable Risk Guaranteescited by 21
  15. 2026Knowing When to Abstain: Medical LLMs Under Clinical Uncertaintycited by 15
  16. 2024Detecting LLM Hallucination Through Layer-wise Information Deficiency: Analysis of Ambiguous Prompts and Unanswerable Questionscited by 9
  17. 2025Domain-Shift-Aware Conformal Prediction for Large Language Modelscited by 6

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Uncertainty Quantification for LLM Agents: A Taxonomy, an Evaluation Protocol, and an Empirical Studycited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025TECP: Token-Entropy Conformal Prediction for LLMs
  2. 2026Conformalized Large Language Models under Configuration Shift
  3. 2025Conformal P-Value in Multiple-Choice Question Answering Tasks with Provable Risk Control
  4. 2025Robust Uncertainty Quantification for Self-Evolving Large Language Models via Continual Domain Pretraining
  5. 2026Is Conformal Factuality for RAG-based LLMs Robust? Novel Metrics and Systematic Insights
  6. 2026Between the Layers Lies the Truth: Uncertainty Estimation in LLMs Using Intra-Layer Local Information Scores
  7. 2026Differentiable Conformal Training for LLM Reasoning Factuality
  8. 2026Does Compression Preserve Uncertainty? A Unified Benchmark for Quantized and Sparse LLMs via Conformal Prediction