# PROVE: Proof-guided Regime-aware Operator Verification for Hallucination Detection in Medical Visual Question Answering

arXiv: https://arxiv.org/abs/2609.29604 · HTML: https://mlsanity.com/paper/2609.29604/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Hallucination mitigation in LVLMs, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 11 of its 29 references that are in the map.
- First posted on arXiv 28 Aug 2026.

## Placed in

- Multimodal LLMs / vision-language models / [Hallucination mitigation in LVLMs](https://mlsanity.com/mllm/hallucination-mitigation-in-lvlms/)

## Builds on (most cited first, 11 of 11)

- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Qwen3-VL Technical Report](https://mlsanity.com/paper/2511.21631/) (2025)
- [SLAKE: A Semantically-Labeled Knowledge-Enhanced Dataset for Medical Visual Question Answering](https://mlsanity.com/paper/2102.09542/) (2021)
- [GREEN: Generative Radiology Report Evaluation and Error Notation](https://mlsanity.com/paper/2405.03595/) (2024)
- [SAC3: Reliable Hallucination Detection in Black-Box Language Models via Semantic-aware Cross-check Consistency](https://mlsanity.com/paper/2311.01740/) (2023)
- [MedHallu: A Comprehensive Benchmark for Detecting Medical Hallucinations in Large Language Models](https://mlsanity.com/paper/2502.14302/) (2025)
- [Pelican: Correcting Hallucination in Vision-LLMs via Claim Decomposition and Program of Thought Verification](https://mlsanity.com/paper/2407.02352/) (2024)
- [Localizing Before Answering: A Hallucination Evaluation Benchmark for Grounded Medical Multimodal LLMs](https://mlsanity.com/paper/2505.00744/) (2025)
- [Combating Multimodal LLM Hallucination via Bottom-Up Holistic Reasoning](https://mlsanity.com/paper/2412.11124/) (2024)
- [CXRAgent: Director-Orchestrated Multi-Stage Reasoning for Chest X-Ray Interpretation](https://mlsanity.com/paper/2510.21324/) (2025)
- [When to Trust the Answer: Question-Aligned Semantic Nearest Neighbor Entropy for Safer Surgical VQA](https://mlsanity.com/paper/2511.01458/) (2025)

## Abstract (arXiv)

In medical visual question answering (VQA), hallucinations of vision-language models (VLMs) may lead to confident but incorrect responses, raising the risk of diagnostic errors. Existing hallucination detection methods uniformly estimate the reliability of VLM outputs from response consistency or visual evidence. However, such uniform verification across questions ignores question-specific characteristics, resulting in missed overconfident errors and false alarms from over-verification. We present PROVE (Proof-guided Regime-aware Operator Verification), a black-box detector that adapts verification strategy to the evidential structure of each question. PROVE classifies questions into three verification regimes based on what kind of visual proof they demand, activates a regime-specific subset of five complementary operators, and adjusts operator importance per question through a lightweight calibration layer conditioned on deterministic question-answer features. PROVE uses question-specific evidence to reweight operators and produce a calibrated risk score. Evaluated on 8048 test samples across three medical VQA benchmarks and four frontier VLMs, PROVE achieves 0.821 AUROC, outperforming the strongest baseline by +0.159, with consistent gains across all models and benchmarks.