Automated paper-to-code reproduction on Paper2CodeBench 2024
4.35Score (ICLR 2024)paper2code + auto-plan & code optimized
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| paper2code + auto-plan & code optimizedLLM=GPT-4.12025.12 | 4.35 | 4.43 | 4.23 | 4.34 | |
| paper2code + auto-code optimizedLLM=GPT-4.12025.12 | 4.28 | 4.39 | 4.18 | 4.29 | |
| paper2code + self-refine in planLLM=GPT-4.12025.12 | 4.14 | 4.26 | 3.75 | 4.05 | |
| paper2code + auto-plan optimizedLLM=GPT-4.12025.12 | 4.01 | 4 | 3.8 | 3.94 | |
| paper2codeLLM=GPT-4.12025.12 | 3.85 | 4.09 | 3.6 | 3.84 |