Automatic experiment reproduction on REPRODUCEBENCH
7.36Method Component ScoreAUTOREPRODUCE
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| AUTOREPRODUCELLM=o3-mini2025.05 | 7.36 | 3.73 | 3.52 | 14.61 | |
| AUTOREPRODUCELLM=Claude-3.5-Sonnet2025.05 | 7.23 | 3.69 | 3.27 | 14.19 | |
| PaperCoderLLM=o3-mini2025.05 | 6.84 | 3.46 | 2.92 | 13.24 | |
| ChatDevLLM=GPT-4o2025.05 | 4.08 | 2.85 | 1.92 | 8.86 |