MIRA: A Bilingual Benchmark for Medical Information Response Audit

• URL: http://arxiv.org/abs/2605.28025v1
• Date: Wed, 27 May 2026 06:28:03 GMT
• Title: MIRA: A Bilingual Benchmark for Medical Information Response Audit
• Abstract summary: We introduce the Medical Information Response Audit (MIRA), a benchmark that assesses whether large language models provide comparable medical information across user-side language, register, and health literacy signals.<n>MIRA contains 4,320 prompts from 60 medically reviewed, low-risk health questions.<n>Across five mainstream LLMs, models answered all medical questions, but responses to low health-literacy signals consistently omitted more key information.<n>A knowledge-guided mitigation prompt reduces information dilution for most models.
• Score: 15.237529171433346
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large language models (LLMs) are increasingly used to provide public-facing health information, yet existing safety evaluations overlook whether responses preserve comparable medical information across different user phrasings of the same question. To address this, we introduce the Medical Information Response Audit (MIRA), a bilingual, controlled benchmark that assesses whether LLMs provide comparable medical information across user-side language, register, and health literacy signals. MIRA contains 4,320 prompts built from 60 medically reviewed, low-risk health questions. Across five mainstream LLMs, models answered all medical questions, but responses to low health-literacy signals consistently omitted more key information, provided fewer concrete next steps, and offered less support for independent judgment. We term this pattern Differential Information Dilution (DID). Language effects are model-specific rather than uniformly worse for non-English prompts. A comparison with 300 real-world health queries provides preliminary evidence of rank-order validity. A knowledge-guided mitigation prompt reduces information dilution for most models, with the largest reductions in underinformative simplification observed for Claude (~8%) and Qwen (~6%).

Related papers

• EchoBench: Benchmarking Sycophancy in Medical Large Vision-Language Models [82.43729208063468]
  Recent benchmarks for medical Large Vision-Language Models (LVLMs) emphasize leaderboard accuracy, overlooking reliability and safety.<n>We study sycophancy -- models' tendency to uncritically echo user-provided information.<n>We introduce EchoBench, a benchmark to systematically evaluate sycophancy in medical LVLMs.
  arXiv  Detail & Related papers  (2025-09-24T14:09:55Z)
• Improving Reliability and Explainability of Medical Question Answering through Atomic Fact Checking in Retrieval-Augmented LLMs [15.61511109105186]
  Large language models (LLMs) exhibit extensive medical knowledge but are prone to hallucinations and inaccurate citations.<n>Current methods, such as Retrieval Augmented Generation, partially address these issues by grounding answers in source documents.<n>We introduce a novel atomic fact-checking framework designed to enhance the reliability and explainability of LLMs.
  arXiv  Detail & Related papers  (2025-05-30T17:33:07Z)
• PerMedCQA: Benchmarking Large Language Models on Medical Consumer Question Answering in Persian Language [0.1747623282473278]
  PerMedCQA is the first Persian-language benchmark for evaluating large language models for medical consumer question answering.<n>We evaluate several state-of-the-art multilingual and instruction-tuned LLMs, utilizing MedJudge, a novel-based evaluation framework driven by an LLM grader.<n>Our results highlight key challenges in multilingual medical QA and provide valuable insights for developing more accurate and context-aware medical assistance systems.
  arXiv  Detail & Related papers  (2025-05-23T19:39:01Z)
• Structured Outputs Enable General-Purpose LLMs to be Medical Experts [50.02627258858336]
  Large language models (LLMs) often struggle with open-ended medical questions.<n>We propose a novel approach utilizing structured medical reasoning.<n>Our approach achieves the highest Factuality Score of 85.8, surpassing fine-tuned models.
  arXiv  Detail & Related papers  (2025-03-05T05:24:55Z)
• Fact or Guesswork? Evaluating Large Language Models' Medical Knowledge with Structured One-Hop Judgments [108.55277188617035]
  Large language models (LLMs) have been widely adopted in various downstream task domains, but their abilities to directly recall and apply factual medical knowledge remains under-explored.<n>We introduce the Medical Knowledge Judgment dataset (MKJ), a dataset derived from the Unified Medical Language System (UMLS), a comprehensive repository of standardized vocabularies and knowledge graphs.<n>Through a binary classification framework, MKJ evaluates LLMs' grasp of fundamental medical facts by having them assess the validity of concise, one-hop statements.
  arXiv  Detail & Related papers  (2025-02-20T05:27:51Z)
• An Empirical Evaluation of Large Language Models on Consumer Health Questions [0.30723404270319693]
  This study evaluates the performance of several Large Language Models (LLMs) on MedRedQA, a dataset of consumer-based medical questions and answers.<n> GPT-4o mini achieved the highest alignment with expert responses according to four out of the five models' judges, while Mistral-7B scored lowest according to three out of five models' judges.
  arXiv  Detail & Related papers  (2024-12-31T01:08:15Z)
• Evaluating Search Engines and Large Language Models for Answering Health Questions [3.8984586307450093]
  This study compares the performance of four popular SEs, seven LLMs, and retrieval-augmented (RAG) variants in answering 150 health-related questions.<n>Findings reveal SEs correctly answer between 50 and 70% of questions, often hindered by many retrieval results not responding to the health question.<n>RAG methods significantly enhance smaller LLMs' effectiveness, improving accuracy by up to 30% by integrating retrieval evidence.
  arXiv  Detail & Related papers  (2024-07-17T10:40:39Z)
• Uncertainty Estimation of Large Language Models in Medical Question Answering [60.72223137560633]
  Large Language Models (LLMs) show promise for natural language generation in healthcare, but risk hallucinating factually incorrect information. We benchmark popular uncertainty estimation (UE) methods with different model sizes on medical question-answering datasets. Our results show that current approaches generally perform poorly in this domain, highlighting the challenge of UE for medical applications.
  arXiv  Detail & Related papers  (2024-07-11T16:51:33Z)
• MedAlign: A Clinician-Generated Dataset for Instruction Following with Electronic Medical Records [60.35217378132709]
  Large language models (LLMs) can follow natural language instructions with human-level fluency. evaluating LLMs on realistic text generation tasks for healthcare remains challenging. We introduce MedAlign, a benchmark dataset of 983 natural language instructions for EHR data.
  arXiv  Detail & Related papers  (2023-08-27T12:24:39Z)
• Medical Misinformation in AI-Assisted Self-Diagnosis: Development of a Method (EvalPrompt) for Analyzing Large Language Models [4.8775268199830935]
  This study aims to assess the effectiveness of large language models (LLMs) as a self-diagnostic tool and their role in spreading healthcare misinformation.<n>We use open-ended questions to mimic real-world self-diagnosis use cases, and perform sentence dropout to mimic realistic self-diagnosis with missing information.<n>The results highlight the modest capabilities of LLMs, as their responses are often unclear and inaccurate.
  arXiv  Detail & Related papers  (2023-07-10T21:28:26Z)
• Self-Verification Improves Few-Shot Clinical Information Extraction [73.6905567014859]
  Large language models (LLMs) have shown the potential to accelerate clinical curation via few-shot in-context learning. They still struggle with issues regarding accuracy and interpretability, especially in mission-critical domains such as health. Here, we explore a general mitigation framework using self-verification, which leverages the LLM to provide provenance for its own extraction and check its own outputs.
  arXiv  Detail & Related papers  (2023-05-30T22:05:11Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.