# Forged Peer Judgments Mislead Multimodal LLM Judge Panels: Source-Blind Anchoring and Panel-Consensus Verification

arXiv: https://arxiv.org/abs/2608.07920 · HTML: https://mlsanity.com/paper/2608.07920/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in VLM benchmarks and visual reasoning, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 7 of its 9 references that are in the map.
- First posted on arXiv 8 Aug 2026.

## Placed in

- Multimodal LLMs / vision-language models / [VLM benchmarks and visual reasoning](https://mlsanity.com/mllm/vlm-benchmarks-and-visual-reasoning/)

## Builds on (most cited first, 7 of 7)

- [RLAIF-V: Open-Source AI Feedback Leads to Super GPT-4V Trustworthiness](https://mlsanity.com/paper/2405.17220/) (2024)
- [VL-RewardBench: A Challenging Benchmark for Vision-Language Generative Reward Models](https://mlsanity.com/paper/2411.17451/) (2024)
- [Nine Judges, Two Effective Votes: Correlated Errors Undermine LLM Evaluation Panels](https://mlsanity.com/paper/2605.29800/) (2026)
- [Fooling the LVLM Judges: Visual Biases in LVLM-Based Evaluation](https://mlsanity.com/paper/2505.15249/) (2025)
- [MM-JudgeBias: A Benchmark for Evaluating Compositional Biases in MLLM-as-a-Judge](https://mlsanity.com/paper/2604.18164/) (2026)
- [Mitigating Perceptual Judgment Bias in Multimodal LLM-as-a-Judge via Perceptual Perturbation and Reward Modeling](https://mlsanity.com/paper/2606.02578/) (2026)
- [On the Adversarial Robustness of Multimodal LLM Judges](https://mlsanity.com/paper/2606.15608/) (2026)

## Abstract (arXiv)

Multimodal LLM judge panels can cross-reference peers, but a quoted peer judgment may itself be untrusted. We expose source-blind anchoring as a text-level attack surface in vision-language model (VLM) panels. Quoting independent visual judgments creates large anchoring gaps (19--26 percentage points) under both self and peer framing. A matched-content, label-only control changes the broken rate by only -0.17pp (95% CI [-0.68,0.35]), showing that the self/peer label itself does not explain the effect. Under our tested construction, deliberately generated, concise wrong quotes overturn originally-correct verdicts 1.5--2.7× more often than naturally occurring wrong peer statements, with bootstrap 95% CIs excluding parity across two datasets and seven VLM judges. Because the two statement populations differ in selection and form, this ratio measures differential damage under the tested attack rather than a provenance-only causal effect. We then introduce panel-consensus verification, which cross-checks a quote against independently collected blind votes. It blocks 84.9% of fabricated attacks, cuts their net harm by 97.5%, and preserves the positive but statistically inconclusive point estimate for genuine peer information under leave-one-out re-verification. These results identify a low-cost attack surface and a concrete defense for safer multimodal collaborative evaluation.