# Seeing Through the Forgery Is Not Enough: Regressive Sycophancy under Fabricated Evidence in Medical Vision-Language Models

arXiv: https://arxiv.org/abs/2610.17989 · HTML: https://mlsanity.com/paper/2610.17989/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Medical vision-language models, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 22 of its 56 references that are in the map.
- First posted on arXiv 13 Oct 2026.

## Placed in

- Multimodal LLMs / vision-language models / [Medical vision-language models](https://mlsanity.com/mllm/medical-vision-language-models/)

## Builds on (most cited first, 22 of 22)

- [Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Prompting](https://mlsanity.com/paper/2305.04388/) (2023)
- [Discovering Language Model Behaviors with Model-Written Evaluations](https://mlsanity.com/paper/2212.09251/) (2022)
- [Quantifying Language Models' Sensitivity to Spurious Features in Prompt Design or: How I learned to start worrying about prompt formatting](https://mlsanity.com/paper/2310.11324/) (2023)
- [Measuring Sycophancy of Language Models in Multi-turn Dialogues](https://mlsanity.com/paper/2505.23840/) (2025)
- [Are You Sure? Challenging LLMs Leads to Performance Drops in The FlipFlop Experiment](https://mlsanity.com/paper/2311.08596/) (2023)
- [The Hawthorne Effect in Reasoning Models: Evaluating and Steering Test Awareness](https://mlsanity.com/paper/2505.14617/) (2025)
- [LLMs Know They're Wrong and Agree Anyway: The Shared Sycophancy-Lying Circuit](https://mlsanity.com/paper/2604.19117/) (2026)
- [When Correct Beliefs Collapse: Epistemic Resilience of LLMs under Clinical Pressure](https://mlsanity.com/paper/2605.23932/) (2026)
- [LURE: Live-Usage Replay Evaluations for Reducing Evaluation Awareness](https://mlsanity.com/paper/2605.26438/) (2026)
- [To See or To Please: Uncovering Visual Sycophancy and Split Beliefs in VLMs](https://mlsanity.com/paper/2603.18373/) (2026)
- [Evaluation Awareness Is Not One Capability: Evidence from Open Language Models](https://mlsanity.com/paper/2606.23583/) (2026)
- [Why LLMs Give In: Conversational Factors and Reasoning Behind Medical Sycophancy](https://mlsanity.com/paper/2608.01017/) (2026)
- [SPINE: Measuring LLM Sycophancy under Sustained Multi-Turn Pressure](https://mlsanity.com/paper/2609.09090/) (2026)
- [Sycophancy Undermines Epistemic Vigilance in Cooperative Vision-Language Tasks](https://mlsanity.com/paper/2607.29585/) (2026)
- [MedPRESS: A Multi-turn Benchmark for Patient-Pressure-Induced Medical Sycophancy in LLMs](https://mlsanity.com/paper/2608.02520/) (2026)
- [Same Semantics, Different Outcome: On the Modality Robustness of Multimodal LLMs under Knowledge Conflict](https://mlsanity.com/paper/2609.00550/) (2026)
- [The Chain Holds, the Answer Folds: Trace-Answer Dissociation in Reasoning Models Under Adversarial Pressure](https://mlsanity.com/paper/2605.29087/) (2026)
- [Position, Not Provenance: Separating Reasoning Mediation from Sycophancy in Medical Vision-Language Models](https://mlsanity.com/paper/2607.27304/) (2026)
- [Looking Again: Measuring Sycophancy in the Reasoning Chains of Multimodal Models Under Pressure](https://mlsanity.com/paper/2608.28623/) (2026)
- [Do Multimodal LLMs See Before They Read? Diagnosing Contextual Sycophancy](https://mlsanity.com/paper/2609.00067/) (2026)
- [GradeTrap: Authority Cues in Images Shift VLM Judgments Despite Explicit Instructions to Ignore Them](https://mlsanity.com/paper/2609.06058/) (2026)
- [Seeing is not Enough: Vision-Language Models Perceive Evidence but Fail to Act](https://mlsanity.com/paper/2609.22588/) (2026)

## Abstract (arXiv)

A medical vision-language model may analyze a given CT scan and correctly detect both kidneys in view. Continuing this conversation, if a user provides a prescription with an opposite direction, medical VLMs often sycophantically choose to realign their diagnosis. In this work, we study regressive sycophancy under fabricated evidence, where a model abandons a correct answer when pushback comes with a fake document. We test four open VLMs on 400 VQA-RAD yes/no questions in 19,200 multi-turn conversations, with and without a fabricated prescription. Without a disclosure that documents are synthetic, the models abandon 80.1% of 6,488 correct answers. Notably, under one script each, the same prescription raises the flip rate of Qwen2.5-VL-7B from 45.9% to 89.6% and lowers that of Qwen3-VL-8B from 77.7% to 50.9%. A model rejected the prescription while still preserving the correct answer in 915 of 3,232 conversations with the prescription (28.3%), and 40% of these flipped in later turns. System-prompt sentences that disclose the documents as synthetic lower one model's flip rate from 77.8% to 47.3%. Thus, we conclude that awareness about a forgery does not prevent a model from adopting its answer.