Competitive Programming on USACO Benchmark
19.9Pass@1Retrieve w/ASM-PL
Evaluation Results
| Method | Links | |
|---|---|---|
| Retrieve w/ASM-PLModel=GPT-4o, Shot-count=1-shot2025.07 | 19.9 | |
| Retrieve w/ASM-NLModel=GPT-4o, Shot-count=1-shot2025.07 | 19.2 | |
| Episodic Retrieval*Model=GPT-4o, Shot-count=1-shot2025.07 | 18.6 | |
| Retrieve w/RandomModel=GPT-4o, Shot-count=1-shot2025.07 | 17.9 | |
| w/o ExemplarModel=GPT-4o, Shot-count=1-shot2025.07 | 17.3 | |
| Retrieve w/StatementModel=GPT-4o, Shot-count=1-shot2025.07 | 16.9 | |
| Retrieve w/ASM-NLModel=GPT-4o-mini, Shot-count=1-shot2025.07 | 13.7 | |
| Retrieve w/ASM-PLModel=GPT-4o-mini, Shot-count=1-shot2025.07 | 13 | |
| Episodic Retrieval*Model=GPT-4o-mini, Shot-count=1-shot2025.07 | 12.4 | |
| Retrieve w/StatementModel=GPT-4o-mini, Shot-count=1-shot2025.07 | 11.4 | |
| Retrieve w/RandomModel=GPT-4o-mini, Shot-count=1-shot2025.07 | 10.4 | |
| w/o ExemplarModel=GPT-4o-mini, Shot-count=1-shot2025.07 | 9.4 |