RS-Neg MCQ - Attribute Negation: leaderboard
Metric: Accuracy (%; pick the one correct caption of four, the distractors mixing affirmative, negated and hybrid descriptions; the 791 items whose negation is attribute-level; zero-shot). Source: arxiv.org. Saturation forecast: Around December 2026. 7 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3 VL 8B Instruct | 58.15 |
| 2 | Qwen 2.5 VL 7B Instruct | 55.25 |
Interactive version: theaggregate.ai/benchmark?slug=rs-neg-mcq-attribute-negation · How It Works · Data refreshed daily, snapshot 2026-09-26.