Crystal Generation on ALM Bench (test)
0.423Application ScoreALM Edit
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| ALM EditN=7 × 10002026.06 | 0.423 | 0.73 | 0.412 | 0.474 | 0.231 | |
| GPT-4o-miniN=7 × 10002026.06 | 0.252 | 0.356 | 0.162 | 0.263 | 0.075 | |
| GPT-4 TurboN=7 × 10002026.06 | 0.224 | 0.254 | 0.09 | 0.168 | 0.035 | |
| GPT-4oN=7 × 10002026.06 | 0.131 | 0.279 | 0.121 | 0.13 | 0.025 |