# Position, Not Provenance: Separating Reasoning Mediation from Sycophancy in Medical Vision-Language Models

arXiv: https://arxiv.org/abs/2607.27304 · HTML: https://mlsanity.com/paper/2607.27304/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Clinical and biomedical applications, one of 17 directions in LLM reasoning.
- Builds on 11 of its 25 references that are in the map.
- First posted on arXiv 29 Jul 2026.

## Placed in

- LLM reasoning / [Clinical and biomedical applications](https://mlsanity.com/lr/clinical-and-biomedical-applications/)
- Multimodal LLMs / vision-language models / [Medical vision-language models](https://mlsanity.com/mllm/medical-vision-language-models/)

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

- [Mistral 7B](https://mlsanity.com/paper/2310.06825/) (2023)
- [Multimodal Chain-of-Thought Reasoning in Language Models](https://mlsanity.com/paper/2302.00923/) (2023)
- [Measuring Faithfulness in Chain-of-Thought Reasoning](https://mlsanity.com/paper/2307.13702/) (2023)
- [MedGemma Technical Report](https://mlsanity.com/paper/2507.05201/) (2025)
- [Explanation-Driven Counterfactual Testing for Faithfulness in Vision-Language Model Explanations](https://mlsanity.com/paper/2510.00047/) (2025)
- [S-Chain: Structured Visual Chain-of-Thought For Medicine](https://mlsanity.com/paper/2510.22728/) (2025)
- [EchoBench: Benchmarking Sycophancy in Medical Large Vision-Language Models](https://mlsanity.com/paper/2509.20146/) (2025)
- [Evaluating Reasoning Faithfulness in Medical Vision-Language Models using Multimodal Perturbations](https://mlsanity.com/paper/2510.11196/) (2025)
- [VFaith: Do Large Multimodal Models Really Reason on Seen Images Rather than Previous Memories?](https://mlsanity.com/paper/2506.11571/) (2025)
- [Benchmarking and Mitigating Sycophancy in Medical Vision Language Models](https://mlsanity.com/paper/2509.21979/) (2025)
- [Medical Context Distorts Decisions in Clinical Vision Language Models](https://mlsanity.com/paper/2605.17436/) (2026)

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

- [Seeing Through the Forgery Is Not Enough: Regressive Sycophancy under Fabricated Evidence in Medical Vision-Language Models](https://mlsanity.com/paper/2610.17989/) (2026)

## Abstract (arXiv)

Medical vision-language models (VLMs) generate chain-of-thought (CoT) reasoning before answering clinical questions, but whether this reasoning causally influences predictions remains unclear. We present CoT-Mediate, a behavioral framework that perturbs a single clinically meaningful attribute within a model's own generated reasoning and measures whether the resulting prediction follows the edited reasoning. Our framework combines a dual-arm protocol comparing re-prompted evidence with prefix-forced continuation, together with a provenance-controlled intervention that varies only the attributed source of identical reasoning to disentangle reasoning mediation from sycophancy. We evaluate LLaVA-Med and MedGemma on 1,000 VQA-RAD samples each. Prefix-forced continuation consistently yields higher mediation faithfulness than re-prompting, while the provenance analysis reveals distinct model-specific deference behaviors. Across both models, removing visual evidence increases reliance on injected reasoning, whereas laterality is the least faithfully tracked clinical attribute. These results show that the mechanism used to inject reasoning substantially affects measured faithfulness and that contextual position, rather than stated provenance, is the primary determinant of whether medical VLMs use their generated reasoning.