Subjective Guidance Quality on Ego4D
3.94Subjective Guidance Quality (1-5)Llama 4-Maverick (PWR)
Evaluation Results
| Method | Links | |
|---|---|---|
| Llama 4-Maverick (PWR)Config=PWR2026.06 | 3.94 | |
| Qwen3.6-VL-27B (PWR)Config=PWR2026.06 | 3.55 | |
| Llama 4-MaverickConfig=FT2026.06 | 2.93 | |
| Qwen3.6-VL-27BConfig=ZS2026.06 | 2.7 | |
| Qwen3-VL-235BConfig=ZS2026.06 | 2.48 | |
| Claude Opus 4.6Config=ZS2026.06 | 2.47 | |
| Gemini 3.1 ProConfig=ZS2026.06 | 2.44 | |
| Llama 4-MaverickConfig=ZS2026.06 | 2.41 | |
| Qwen3.6-VL-27BConfig=FT2026.06 | 2.39 |