Robot Failure Explanation on RoboFail
0.628Coherence Score (CS)Gemini 2.5 Pro
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Gemini 2.5 Pro2025.06 | 0.628 | 0.342 | 0.55 | |
| AHA-13B2025.06 | 0.471 | 0.28 | 0.465 | |
| RoboFAC-7B2025.06 | 0.452 | 0.137 | 0.133 |
| Method | Links | |||
|---|---|---|---|---|
| Gemini 2.5 Pro2025.06 | 0.628 | 0.342 | 0.55 | |
| AHA-13B2025.06 | 0.471 | 0.28 | 0.465 | |
| RoboFAC-7B2025.06 | 0.452 | 0.137 | 0.133 |