Multi-step Operation Sequence Prediction on GUIOdyssey
46.18PrecisionTeach VLM
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Teach VLM2026.06 | 46.18 | 46.78 | 46.48 | 53.81 | |
| Doubao-Seed-1.82026.06 | 27.46 | 25.46 | 26.42 | 72.54 | |
| Qwen3-VL-8B2026.06 | 26.4 | 20.03 | 22.78 | 73.6 | |
| Doubao-Seed-2.02026.06 | 23.9 | 22.46 | 23.16 | 76.1 | |
| GPT-4o2026.06 | 18.46 | 16.69 | 17.53 | 81.54 | |
| Qwen3.5-9B2026.06 | 5.18 | 8.77 | 6.51 | 94.82 |