mlsanityData to 14 Oct 2026

Geometry-Calibrated Conformal Abstention for Language Models

arXiv 2604.27914 ↗First posted 30 Apr 2026cs.CL cs.LGcited by 5

Not on the map. Why?

Abstract (arXiv)

When language models lack relevant knowledge for a given query, they frequently generate plausible responses that can be hallucinations, rather than admitting being agnostic about the answer. Retraining models to reward admitting ignorance can lead to overly conservative behaviors and poor generalization due to scarce evaluation benchmarks. We propose a post hoc framework, Conformal Abstention (CA), adapted from conformal prediction (CP) to determine whether to abstain from answering a query. CA provides finite-sample guarantees on both the probability of participation (i.e., not abstaining) and the probability that the generated response is correct. Importantly, the abstention decision relies on prediction confidence rather than the non-conformity scores used in CP, which are intractable for open-ended generation. To better align prediction confidence with the model's ignorance, we introduce a calibration strategy using representation geometry within the model to measure knowledge involvement in shaping the response. Experiments demonstrate that we improve selective answering significantly with 75 percent conditional correctness.

Key facts. Cited by 5 papers in the map (data through 14 Oct 2026). Builds on 37 of its 54 references that are in the map. First posted on arXiv 30 Apr 2026.

Builds on

37 of 54 references.

References · 37

  1. 2017Attention Is All You Needcited by 60,361
  2. 2021Training Verifiers to Solve Math Word Problemscited by 10,109
  3. 2019BERTScore: Evaluating Text Generation with BERTcited by 5,512
  4. 2016XGBoost: A Scalable Tree Boosting Systemcited by 5,166
  5. 2022Language Models (Mostly) Know What They Knowcited by 1,856
  6. 2018CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledgecited by 1,822
  7. 2023A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questionscited by 1,601
  8. 2018A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attackscited by 1,485
  9. 2023Semantic Uncertainty: Linguistic Invariances for Uncertainty Estimation in Natural Language Generationcited by 819
  10. 2019How Contextual are Contextualized Word Representations? Comparing the Geometry of BERT, ELMo, and GPT-2 Embeddingscited by 784
  11. 2017Crowdsourcing Multiple Choice Science Questionscited by 765
  12. 2020Uncertainty Sets for Image Classifiers using Conformal Predictioncited by 366
25 more
  1. 2024INSIDE: LLMs' Internal States Retain the Power of Hallucination Detectioncited by 323
  2. 2023Robots That Ask For Help: Uncertainty Alignment for Large Language Model Plannerscited by 308
  3. 2020Unsupervised Quality Estimation for Neural Machine Translationcited by 211
  4. 2022Out-of-Distribution Detection and Selective Generation for Conditional Language Modelscited by 197
  5. 2023Conformal Language Modelingcited by 170
  6. 2023Shifting Attention to Relevance: Towards the Predictive Uncertainty Quantification of Free-Form Large Language Modelscited by 150
  7. 2023Conformal Prediction with Large Language Models for Multi-Choice Question Answeringcited by 139
  8. 2024Language Models with Conformal Factuality Guaranteescited by 131
  9. 2024Know Your Limits: A Survey of Abstention in Large Language Modelscited by 121
  10. 2024Benchmarking Uncertainty Quantification Methods for Large Language Models with LM-Polygraphcited by 105
  11. 2024Chain of Thoughtlessness? An Analysis of CoT in Planningcited by 103
  12. 2024A Comprehensive Survey on Process-Oriented Automatic Text Summarization with Exploration of LLM-Based Methodscited by 102
  13. 2025Unveiling Privacy Risks in LLM Agent Memorycited by 101
  14. 2023Enhancing Uncertainty-Based Hallucination Detection with Stronger Focuscited by 99
  15. 2022Measuring the Mixing of Contextual Information in the Transformercited by 65
  16. 2024ConU: Conformal Uncertainty in Large Language Models with Correctness Coverage Guaranteescited by 65
  17. 2024Reasoning Aware Self-Consistency: Leveraging Reasoning Paths for Efficient LLM Samplingcited by 63
  18. 2024Determinants of LLM-assisted Decision-Makingcited by 56
  19. 2024Conformal Prediction for Natural Language Processing: A Surveycited by 53
  20. 2024Towards Scientific Discovery with Generative AI: Progress, Opportunities, and Challengescited by 41
  21. 2024Uncertainty-Aware Evaluation for Vision-Language Modelscited by 29
  22. 2025Analyzing Uncertainty of LLM-as-a-Judge: Interval Evaluations with Conformal Predictioncited by 27
  23. 2024Sample then Identify: A General Framework for Risk Control and Assessment in Multimodal Large Language Modelscited by 19
  24. 2025Token-Level Density-Based Uncertainty Quantification Methods for Eliciting Truthfulness of Large Language Modelscited by 18
  25. 2025Hallucination Detection and Evaluation of Large Language Modelcited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 5.

5 papers.

  1. 2026Autonomous Research Agents: A Survey of AI Scientists and the Verification Gapcited by 9
  2. 2026Safe to Stop? Risk-Constrained Stopping for Sequential Clinical Diagnosis Agentscited by 1
  3. 2026Can a Model Catch Its Own Hallucinations for Free?: Label-Free Doubt Signals Hold Their Own Against a Labelled Dataset for Abstentioncited by 0
  4. 2026Uncertainty Quantification for LLM Agents: A Taxonomy, an Evaluation Protocol, and an Empirical Studycited by 0
  5. 2026Certified Against Which Oracle? Execution Labels Set the Reported Risk of Conformal Abstention for Text-to-SQLcited by 0