MM-JudgeBias - Unnecessary Image: leaderboard
MM-JudgeBias (2026) tests multimodal LLM judges for compositional bias: the judge scores image, query and response triplets from 1 to 10 before and after a controlled perturbation. Bias-Conformity from 0 to 1, the stability of the score when an unrelated image is added to a text-only task; higher is better.
Metric: Bias-Conformity (0 to 1) on MM-JudgeBias (about 200 image-query-response triplets per bias type, 1,804 in all, each judged on a 1-10 scale before and after a controlled perturbation; mean of three runs): mean score invariance (1 minus the absolute score change over its largest possible value) when an unrelated image is added to a text-only task; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 26 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | LLaVA-Critic-72B | 0.974 | ||
| 2 | GPT-4.1 Mini | 0.972 | #789 | |
| 3 | GPT-5.1 | High | 0.961 | #240 |
| 4 | Gemini 2.0 Flash Lite | 0.958 | #1050 | |
| 5 | GPT-5 Mini | High | 0.954 | #364 |
| 6 | Qwen 3 VL 8B Instruct | 0.952 | #1029 | |
| 7 | O4 Mini | High | 0.95 | #368 |
| 8 | Gemini 3 Pro | High | 0.947 | #145 |
| 9 | Gemini 2.5 Flash | 0.944 | #463 | |
| 10 | Claude Haiku 4.5 | Thinking | 0.943 | #583 |
| 11 | Gemini 2.5 Flash Lite | Thinking | 0.94 | #964 |
| 12 | O3 | High | 0.937 | #237 |
| 13 | LLaVA-Critic-7B | 0.929 | ||
| 14 | Qwen 3 VL 8B (Thinking) | 0.927 | #910 | |
| 15 | Claude Opus 4.5 | Thinking | 0.925 | #160 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=mm-judgebias-unnecessary-image · How It Works · Data refreshed daily, snapshot 2026-11-21.