RoadSafe365 - Dense Captioning: leaderboard
Metric: METEOR (0-1) of the model's detailed description of each accident video (entities, motion and interactions) against the annotated reference caption, on the 700-clip test set; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 11 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | InternVL3-8B | 0.32 | #606 |
| 2 | Qwen 2.5 VL 7B Instruct | 0.27 | #643 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=roadsafe365-dense-captioning · How It Works · Data refreshed daily, snapshot 2026-10-11.