ProactEgoMem - Sci IC - Intervention Start@0.2: leaderboard
ProactEgoMem Sci: egocentric video proactive assistance; IC object identification,ST state transitions,LT location tracking. Notification and tool-call intervention levels,matching includes semantic judging at temperature 0. Start@0.2 uses matched actions; mIoU counts unmatched opportunities as zero. Backbone effort unspecified; conditional timing errors/SD omitted. Score (%).
Metric: Score (%). Source: arxiv.org. Saturation forecast: Around January 2027. 13 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3.8 27B | 46.2 | #118 |
| 2 | Gemini 2.5 Flash | 36.4 | #242 |
| 3 | GPT-5.4 Mini | 20 | #212 |
| 4 | Gemma 4 31B | 13.6 | #200 |
| 5 | Qwen 3.6 27B | 12.5 | #173 |
| 6 | Qwen 3.5 9B | 0 | #384 |
| 7 | Inkling Small | 0 | #160 |
No result here: #4 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=proactegomem-sci-ic-intervention-start-0-2 · How It Works · Data refreshed daily, snapshot 2026-10-23.