mlsanityData to 14 Oct 2026

Mind the Ambiguity: Aleatoric Uncertainty Quantification in LLMs for Safe Medical Question Answering

arXiv 2601.17284 ↗First posted 24 Jan 2026cs.CL cs.AIcited by 2

Not on the map. Why?

Abstract (arXiv)

The deployment of Large Language Models in Medical Question Answering is severely hampered by ambiguous user queries, a significant safety risk that demonstrably reduces answer accuracy in high-stakes healthcare settings. In this paper, we formalize this challenge by linking input ambiguity to aleatoric uncertainty (AU), which is the irreducible uncertainty arising from underspecified input. To facilitate research in this direction, we construct CV-MedBench, the first benchmark designed for studying input ambiguity in Medical QA. Using this benchmark, we analyze AU from a representation engineering perspective, revealing that AU is linearly encoded in LLM's internal activation patterns. Leveraging this insight, we introduce a novel AU-guided "Clarify-Before-Answer" framework, which incorporates AU-Probe - a lightweight module that detects input ambiguity directly from hidden states. Unlike existing uncertainty estimation methods, AU-Probe requires neither LLM fine-tuning nor multiple forward passes, enabling an efficient mechanism to proactively request user clarification and significantly enhance safety. Extensive experiments across four open LLMs demonstrate the effectiveness of our QA framework, with an average accuracy improvement of 9.48% over baselines. Our framework provides an efficient and robust solution for safe Medical QA, strengthening the reliability of health-related applications. The code is available at https://github.com/yaokunliu/AU-Med.git, and the CV-MedBench dataset is released on Hugging Face at https://huggingface.co/datasets/yaokunl/CV-MedBench.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 21 of its 38 references that are in the map. First posted on arXiv 24 Jan 2026.

Builds on

21 of 38 references.

References · 21

  1. 2017On Calibration of Modern Neural Networkscited by 4,960
  2. 2022Language Models (Mostly) Know What They Knowcited by 1,856
  3. 2022Large Language Models Encode Clinical Knowledgecited by 1,830
  4. 2020What Disease does this Patient Have? A Large-scale Open Domain Question Answering Dataset from Medical Examscited by 1,356
  5. 2023Representation Engineering: A Top-Down Approach to AI Transparencycited by 1,336
  6. 2024Refusal in Language Models Is Mediated by a Single Directioncited by 1,032
  7. 2019Aleatoric and Epistemic Uncertainty in Machine Learning: An Introduction to Concepts and Methodscited by 913
  8. 2019Deep Ensembles: A Loss Landscape Perspectivecited by 443
  9. 2020AmbigQA: Answering Ambiguous Open-domain Questionscited by 390
  10. 2024Semantic Entropy Probes: Robust and Cheap Hallucination Detection in LLMscited by 259
  11. 2020Unsupervised Quality Estimation for Neural Machine Translationcited by 211
  12. 2023Decomposing Uncertainty for Large Language Models through Input Clarification Ensemblingcited by 114
9 more
  1. 2020Bayesian Deep Ensembles via the Neural Tangent Kernelcited by 90
  2. 2022Mutual Information Alleviates Hallucinations in Abstractive Summarizationcited by 68
  3. 2024MedExQA: Medical Question Answering Benchmark with Multiple Explanationscited by 60
  4. 2024Distinguishing the Knowable from the Unknowable with Language Modelscited by 47
  5. 2025The Illusion of Certainty: Uncertainty Quantification for LLMs Fails under Ambiguitycited by 17
  6. 2025Efficient Hallucination Detection for LLMs Using Uncertainty-Aware Attention Headscited by 16
  7. 2023Fine-Grained Classification with Noisy Labelscited by 15
  8. 2025The challenge of uncertainty quantification of large language models in medicinecited by 14
  9. 2025Aligned Probing: Relating Toxic Behavior and Model Internalscited by 5

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026A knowledge-guided agentic framework for mitigating patient-context ambiguity in health queriescited by 0
  2. 2026PropUQ-MAS: Propagation-Aware Uncertainty Quantification for LLM Multi-Agent Systemscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Knowing When to Abstain: Medical LLMs Under Clinical Uncertainty
  2. 2026CLEAR: Revealing How Noise and Ambiguity Degrade Reliability in LLMs for Medicine
  3. 2025Mind the Gap: Benchmarking LLM Uncertainty and Calibration with Specialty-Aware Clinical QA and Reasoning-Based Behavioural Features
  4. 2025Tools in the Loop: Quantifying Uncertainty of LLM Question Answering Systems That Use Tools
  5. 2024MAQA: Evaluating Uncertainty Quantification in LLMs Regarding Data Uncertainty
  6. 2026From Answers to Interpretations: Rethinking Ambiguity-Induced Aleatoric Uncertainty Estimation in LLMs
  7. 2025Trustworthy Medical Question Answering: An Evaluation-Centric Survey
  8. 2026When Quantization Preserves Accuracy but Not Evidence: Explanation-Aware Post-Training Quantization for Medical LLMs