CAVSR (6 Categories): leaderboard
CAVSR (2026) child-oriented risk review of 605 real AI-generated short videos annotated by child-education experts, scored over the six top-level risk categories. Micro-F1 of the multi-label predictions in percent, higher is better.
Metric: Micro-F1 (%; multi-label child-risk recognition over six top-level risk categories on 605 expert-annotated AI-generated videos from TikTok, RedNote and YouTube; the model reviews each video directly). Source: arxiv.org. Saturation forecast: Around February 2027. 9 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Flash | 75.16 | #164 |
| 2 | GPT-4o | 59.01 | #741 |
| 3 | Qwen 2.5 VL 7B Instruct | 31.89 | #1833 |
| 4 | Qwen 2 VL 7B | 30.01 | #2261 |
| 5 | Qwen 2 VL 2B | 12.6 | #6284 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=cavsr-6-categories · How It Works · Data refreshed daily, snapshot 2026-11-13.