MM-JudgeBias - Response Dominance: 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-Deviation from 0 to 1, the normalized score drop when both the query and the image are removed; higher is better.
Metric: Bias-Deviation (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 drop of the judge's score, normalized by the largest possible drop, when both the query and the image are replaced with null inputs; 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 | Gemini 3 Pro | High | 0.988 | #145 |
| 2 | Gemini 2.5 Pro | 0.978 | #276 | |
| 3 | Claude Opus 4.5 | Thinking | 0.917 | #160 |
| 4 | Qwen 3 VL 8B Instruct | 0.913 | #1029 | |
| 5 | Gemini 2.5 Flash Lite | Thinking | 0.89 | #964 |
| 6 | Claude Sonnet 4.5 | Thinking | 0.831 | #261 |
| 7 | Qwen 3 VL 30B A3B (Thinking) | 0.806 | #719 | |
| 8 | Gemini 2.5 Flash | 0.743 | #463 | |
| 9 | Claude Haiku 4.5 | Thinking | 0.735 | #583 |
| 10 | Qwen 3 VL 30B A3B Instruct | 0.642 | #895 | |
| 11 | O3 | High | 0.596 | #237 |
| 12 | Gemini 2.0 Flash Lite | 0.57 | #1050 | |
| 13 | O4 Mini | High | 0.518 | #368 |
| 14 | LLaVA-Critic-7B | 0.42 | ||
| 15 | Qwen 3 VL 8B (Thinking) | 0.412 | #910 |
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-response-dominance · How It Works · Data refreshed daily, snapshot 2026-11-21.