EgoArgus (Assistance Decision): leaderboard
EgoArgus (Tang, Liu and Yen, 2026) evaluates vision-language models as egocentric assistants: from a first-person video, a dialogue and its timestamp, the model decides whether help is needed in 789 VISTA episodes. Assistance-decision F1 in percent over all scenarios, including cases where no help is needed; higher is better.
Metric: Assistance-decision F1 (%; whether help is required, from the egocentric video, dialogue and dialogue timestamp of 789 VISTA episodes, all six scenarios including no-assistance cases; API models get the full video, open models sampled frames). Source: arxiv.org. Saturation forecast: Around February 2027. 7 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Flash Lite | 81.17 | #363 |
| 2 | Qwen 3.5 Plus (2026-02-15) | 72.32 | #178 |
| 3 | Molmo2-8B | 53.75 | #1488 |
| 4 | Qwen 3.5 2B | 19.2 | #4849 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=egoargus-assistance-decision · How It Works · Data refreshed daily, snapshot 2026-11-12.