# MedFabric: Gold Evidence Hides the Difficulty of Word-Level Medical Fabrication Detection

arXiv: https://arxiv.org/abs/2605.04180 · HTML: https://mlsanity.com/paper/2605.04180/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 15 of its 28 references that are in the map.
- First posted on arXiv 5 May 2026.

## Builds on (most cited first, 15 of 15)

- [TruthfulQA: Measuring How Models Mimic Human Falsehoods](https://mlsanity.com/paper/2109.07958/) (2021)
- [MiniLM: Deep Self-Attention Distillation for Task-Agnostic Compression of Pre-Trained Transformers](https://mlsanity.com/paper/2002.10957/) (2020)
- [PubMedQA: A Dataset for Biomedical Research Question Answering](https://mlsanity.com/paper/1909.06146/) (2019)
- [Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Models](https://mlsanity.com/paper/2506.05176/) (2025)
- [Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preference](https://mlsanity.com/paper/2403.04132/) (2024)
- [Retrieval Augmentation Reduces Hallucination in Conversation](https://mlsanity.com/paper/2104.07567/) (2021)
- [The Internal State of an LLM Knows When It's Lying](https://mlsanity.com/paper/2304.13734/) (2023)
- [ChatDoctor: A Medical Chat Model Fine-Tuned on a Large Language Model Meta-AI (LLaMA) Using Medical Domain Knowledge](https://mlsanity.com/paper/2303.14070/) (2023)
- [Self-Play Preference Optimization for Language Model Alignment](https://mlsanity.com/paper/2405.00675/) (2024)
- [Med-HALT: Medical Domain Hallucination Test for Large Language Models](https://mlsanity.com/paper/2307.15343/) (2023)
- [MEDEC: A Benchmark for Medical Error Detection and Correction in Clinical Notes](https://mlsanity.com/paper/2412.19260/) (2024)
- [The Factual Inconsistency Problem in Abstractive Text Summarization: A Survey](https://mlsanity.com/paper/2104.14839/) (2021)
- [Retrieving Supporting Evidence for Generative Question Answering](https://mlsanity.com/paper/2309.11392/) (2023)
- [KnowHalu: Hallucination Detection via Multi-Form Knowledge Based Factual Checking](https://mlsanity.com/paper/2404.02935/) (2024)
- [Zero-resource Hallucination Detection for Text Generation via Graph-based Contextual Knowledge Triples Modeling](https://mlsanity.com/paper/2409.11283/) (2024)

## Abstract (arXiv)

Large language models fabricate in medicine, producing fluent statements that are factually wrong, so reliable fabrication detection is a prerequisite for clinical deployment. Reported progress on this task is inflated by two evaluation artifacts: an authorship-style shortcut, where human-written ground truths are paired with LLM-written hallucinations so detectors key on writing style rather than facts, and the provision of gold evidence at test time. A benchmark that tests factual reasoning must therefore remove the style shortcut, ground every fabrication in a real retrievable passage, and score detectors across the range of evidence quality faced in deployment. We build MedFabric to these requirements, a benchmark of 646 word-level medical fabrications, each paired with a ground truth that shares its LLM authorship and near-identical surface form (median ROUGE-L 0.95). On MedFabric the task is unsolved: expert clinicians reach only 53.3% macro F1 and no detector family clears about 60% without gold evidence. Our central finding is that detection is governed by evidence correctness rather than fabrication subtlety, since a strong LLM scores 91% with the gold passage but falls to 35%, below its own no-evidence baseline, under a wrong one, a pattern that holds on two benchmarks at two model scales. The failure is actionable: a retrieval-confidence gate that abstains on low-confidence evidence raises macro F1 from 61% to 74%, and we release MedFabric, all code, and every baseline.