EgoIntrospect - Valuable Interaction Identification: leaderboard
Which of four proactive assistant requests the wearer accepted for a clip, asked on egocentric smart-glasses clips annotated by the wearers themselves (EgoIntrospect, 2026); native audio and video models hear the clip audio and the others read a transcript. Accuracy in percent on 1,161 four-choice questions; higher is better.
Metric: Accuracy (%) on 1,161 four-choice questions asking which of four proactive assistant requests the wearer accepted for a clip (distractors are requests the wearer did not accept), fixation-overlaid egocentric clips from 15 held-out participants with the participant profile; native audio-video models hear the clip audio, the other models read its transcript; temperature 0 (0.6 for InternVL3.5 thinking mode); higher is better. Source: arxiv.org. Saturation forecast: Around 2029. 19 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 55.81 | #99 |
| 2 | Kimi K2.5 | 54.78 | #266 |
| 3 | Gemini 3 Flash (Preview) | 52.8 | #152 |
| 4 | Qwen 3.5 397B A17B | 51.42 | #264 |
| 5 | Qwen 3.5 Omni Plus | 49.96 | #254 |
| 6 | Qwen 3 VL 8B Instruct | 47.63 | #1029 |
| 7 | Qwen 3 Omni 30B A3B (Thinking) | 47.29 | |
| 8 | Qwen3 Omni 30B A3B Instruct | 46.34 | #930 |
| 9 | Qwen 3 VL 8B (Thinking) | 46.17 | #910 |
| 10 | GPT-4o | 43.84 | #751 |
| 11 | InternVL2.5-8B | 43.58 | #2370 |
| 12 | MiniCPM-V-4.5 | 43.32 | #1441 |
| 13 | InternVL3-8B | 42.89 | #1782 |
| 14 | InternVL3.5-8B-Instruct | 41.69 | #1891 |
| 15 | Qwen 2.5 VL 7B Instruct | 41.09 | #1902 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=egointrospect-valuable-interaction-identification · How It Works · Data refreshed daily, snapshot 2026-11-21.