EgoArgus (Assistance Decision) - Text-Grounded: leaderboard
Metric: Assistance-decision F1 (%; whether help is required, from the egocentric video, dialogue and dialogue timestamp of 789 VISTA episodes, text-grounded scenario: the dialogue decides and the video is irrelevant; API models get the full video, open models sampled frames). Source: arxiv.org. Saturation forecast: Estimated already saturated. 7 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Flash Lite | 94.63 |
| 2 | MiMo-V2-Omni | 70.31 |
| 3 | Qwen 3.5 Plus (2026-02-15) | 44.44 |
| 4 | Qwen 3.5 2B | 22.03 |
| 5 | Molmo2-8B | 12.51 |
Interactive version: theaggregate.ai/benchmark?slug=egoargus-assistance-decision-text-grounded · How It Works · Data refreshed daily, snapshot 2026-09-26.