Autoformalization of Linear Programs on NL4LP
97.22Accuracy (%)AutoOR 8B
Evaluation Results
| Method | Links | |
|---|---|---|
| AutoOR 8Bevaluation_protocol=pass@12026.04 | 97.22 | |
| AutoORSize=8B2026.04 | 97.2 | |
| Gemini 2.5 Proevaluation_protocol=pass@12026.04 | 92.59 | |
| Gemini 3 Proevaluation_protocol=pass@12026.04 | 91.67 | |
| OR-R1RLSize=8B2026.04 | 84.6 | |
| OR-R1SFTSize=8B2026.04 | 82.9 | |
| Qwen3-8Bevaluation_protocol=pass@12026.04 | 80.56 | |
| ORLMSize=8B2026.04 | 76.4 |