Interactive step-by-step task guidance on Qualcomm Interactive Cooking (main set)
23.1IC AccuracyGemini-2.5-Flash
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| Gemini-2.5-FlashZero-shot=true2025.11 | 23.1 | 1 | 22 | 2 | 41 | 34.2 | |
| Qwen2.5-VL-7BZero-shot=true2025.11 | 18.9 | 18 | 1 | 2 | 29.9 | 21.9 | |
| Qwen2-VL-7BZero-shot=true2025.11 | 6.3 | 2 | 69 | 5 | 37.7 | 25.6 | |
| Video-LLaVAZero-shot=true2025.11 | 2 | 0 | 0 | 0 | 0 | 0 | |
| VideoLLaMA3-7BZero-shot=true2025.11 | 1.8 | 0 | 0 | 0 | 0 | 0 | |
| Video-ChatGPTZero-shot=true2025.11 | 1.6 | 0 | 0 | 0 | 0 | 0 | |
| VideoChat2Zero-shot=true2025.11 | 1.6 | 0 | 0 | 0 | 0 | 0 | |
| LLaVA-NeXTZero-shot=true2025.11 | 1.4 | 0 | 0 | 0 | 0 | 0 | |
| Videollm-onlineZero-shot=true2025.11 | 0.03 | 2 | 98 | 4 | 33.2 | 24.8 |