Cross-domain Demo-to-Code on Real-world Medium-Complexity Demonstrations
87.5Success Rate (SR)NESYCR
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| NESYCRComplexity Level=Medium-Complexity, Cross-domain factor=Combination2026.03 | 87.5 | 98.21 | 24.49 | |
| NESYCR2026.03 | 75 | 90.56 | 10.57 | |
| GPT4V-RoboticsComplexity Level=Medium-Complexity, Cross-domain factor=Combination2026.03 | 50 | 75 | 0 | |
| StatlerComplexity Level=Medium-Complexity, Cross-domain factor=Combination2026.03 | 50 | 67.86 | 42.86 | |
| MoReVQA2026.03 | 43.33 | 75 | 25.17 | |
| GPT4V-Robotics2026.03 | 41.67 | 68.89 | 43.18 | |
| Statler2026.03 | 41.67 | 71.67 | 49.07 | |
| Critic-V2026.03 | 35 | 64.44 | 24.97 | |
| Demo2Code2026.03 | 25 | 51.11 | 10.67 | |
| LLM-DM2026.03 | 20 | 47.78 | 77.13 | |
| Demo2CodeComplexity Level=Medium-Complexity, Cross-domain factor=Combination2026.03 | 0 | 25 | — |