Generative Operational Planning on ACSE verification environment (ACTS)
61Overall Success RateMPHA
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| MPHA2026.05 | 61 | 60 | 62 | 60 | 0.84 | 12.63 | 81.18 | |
| Gemini 3.1 Pro2026.05 | 49.33 | 48 | 48 | 50.67 | 0.75 | 12.61 | 82.04 | |
| DeepSeek-V32026.05 | 48.67 | 44 | 44 | 53.33 | 1.36 | 14.43 | 89.34 | |
| GLM-52026.05 | 13.33 | 0 | 26.67 | 0 | 2.25 | 17.56 | 110.72 | |
| Qwen3-Max-Thinking2026.05 | 2.67 | 1.33 | 4 | 1.33 | 2.29 | 20.14 | 127.8 |