# Auditing Evidence Use in Medical LLM Diagnosis

arXiv: https://arxiv.org/abs/2607.20848 · HTML: https://mlsanity.com/paper/2607.20848/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 32 of its 46 references that are in the map.
- First posted on arXiv 23 Jul 2026.

## Builds on (most cited first, 30 of 32)

- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [A Unified Approach to Interpreting Model Predictions](https://mlsanity.com/paper/1705.07874/) (2017)
- ["Why Should I Trust You?": Explaining the Predictions of Any Classifier](https://mlsanity.com/paper/1602.04938/) (2016)
- [Axiomatic Attribution for Deep Networks](https://mlsanity.com/paper/1703.01365/) (2017)
- [TruthfulQA: Measuring How Models Mimic Human Falsehoods](https://mlsanity.com/paper/2109.07958/) (2021)
- [Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language models](https://mlsanity.com/paper/2206.04615/) (2022)
- [Large Language Models Encode Clinical Knowledge](https://mlsanity.com/paper/2212.13138/) (2022)
- [Understanding Black-box Predictions via Influence Functions](https://mlsanity.com/paper/1703.04730/) (2017)
- [Datasheets for Datasets](https://mlsanity.com/paper/1803.09010/) (2018)
- [What Disease does this Patient Have? A Large-scale Open Domain Question Answering Dataset from Medical Exams](https://mlsanity.com/paper/2009.13081/) (2020)
- [Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)](https://mlsanity.com/paper/1711.11279/) (2017)
- [PubMedQA: A Dataset for Biomedical Research Question Answering](https://mlsanity.com/paper/1909.06146/) (2019)
- [Qwen2.5-Coder Technical Report](https://mlsanity.com/paper/2409.12186/) (2024)
- [Sanity Checks for Saliency Maps](https://mlsanity.com/paper/1810.03292/) (2018)
- [Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Prompting](https://mlsanity.com/paper/2305.04388/) (2023)
- [Model Cards for Model Reporting](https://mlsanity.com/paper/1810.03993/) (2018)
- [Beyond Accuracy: Behavioral Testing of NLP models with CheckList](https://mlsanity.com/paper/2005.04118/) (2020)
- [Attention is not Explanation](https://mlsanity.com/paper/1902.10186/) (2019)
- [Capabilities of GPT-4 on Medical Challenge Problems](https://mlsanity.com/paper/2303.13375/) (2023)
- [Towards Faithfully Interpretable NLP Systems: How should we define and evaluate faithfulness?](https://mlsanity.com/paper/2004.03685/) (2020)
- [ERASER: A Benchmark to Evaluate Rationalized NLP Models](https://mlsanity.com/paper/1911.03429/) (2019)
- [Learning the Difference that Makes a Difference with Counterfactually-Augmented Data](https://mlsanity.com/paper/1909.12434/) (2019)
- [Is Attention Interpretable?](https://mlsanity.com/paper/1906.03731/) (2019)
- [BioMistral: A Collection of Open-Source Pretrained Large Language Models for Medical Domains](https://mlsanity.com/paper/2402.10373/) (2024)
- [The Shapley Taylor Interaction Index](https://mlsanity.com/paper/1902.05622/) (2019)
- [emrQA: A Large Corpus for Question Answering on Electronic Medical Records](https://mlsanity.com/paper/1809.00732/) (2018)
- [Explaining Explanations: Axiomatic Feature Interactions for Deep Networks](https://mlsanity.com/paper/2002.04138/) (2020)
- [Med42-v2: A Suite of Clinical LLMs](https://mlsanity.com/paper/2408.06142/) (2024)
- [DDXPlus: A New Dataset For Automatic Medical Diagnosis](https://mlsanity.com/paper/2205.09148/) (2022)
- [Detecting Statistical Interactions from Neural Network Weights](https://mlsanity.com/paper/1705.04977/) (2017)

## Built on by (most cited first, 1 of 1)

- [From Hindsight to Foresight: Dissecting Clinical Evidence Acquisition in Large Language Models](https://mlsanity.com/paper/2610.18605/) (2026)

## Abstract (arXiv)

Medical LLMs are often evaluated by whether they select the correct diagnosis, but diagnostic accuracy alone does not show whether the model used the case evidence appropriately. We present a behavioral audit of evidence use in medical diagnosis. For each case, we decompose patient information into evidence units, score candidate diagnoses under controlled evidence subsets, and mine low-order interactions in diagnostic margins. Because medical evidence is diagnosis-relative, the audit separates interaction discovery from failure assignment: large or negative interactions can reflect plausible differential diagnosis, while suspicious interactions require robustness checks and clinical review. We evaluate five open-weight LLMs on DDXPlus, CupCase, and MedCase. Across datasets, faithful support and differential conflict or cancellation account for most interaction strength, showing that many evidence interactions are clinically plausible rather than failures. In a DDXPlus-focused blinded five-reviewer 130-item enriched review sample, invalid or shortcut-like cases concentrate in negated or absent findings and clinically local evidence. These results show that accuracy can hide candidate evidence-use failures and motivate role-aware audits for medical LLM evaluation.