Navigation Reasoning on BBH-Navigate (test)
98AccuracyUPA
Evaluation Results
| Method | Links | |
|---|---|---|
| UPAVenue=-, Execution Engine=GPT-4o-mini2026.01 | 98 | |
| PromptBreederVenue=ICML 24, Execution Engine=GPT-4o-mini2026.01 | 96.3 | |
| SPOVenue=EMNLP 26, Execution Engine=GPT-4o-mini2026.01 | 96.3 | |
| OPROVenue=ICLR 24, Execution Engine=GPT-4o-mini2026.01 | 95.8 | |
| PromptAgentVenue=ICLR 24, Execution Engine=GPT-4o-mini2026.01 | 95.7 | |
| RaRVenue=arXiv 23, Execution Engine=GPT-4o-mini2026.01 | 93.5 | |
| Step-BackVenue=ICLR 24, Execution Engine=GPT-4o-mini2026.01 | 93.5 | |
| APEVenue=ICLR 23, Execution Engine=GPT-4o-mini2026.01 | 92.5 | |
| IOVenue=-, Execution Engine=GPT-4o-mini2026.01 | 91.3 | |
| TextGradVenue=Nature 25, Execution Engine=GPT-4o-mini2026.01 | 91.3 | |
| CoTVenue=NeurIPS 22, Execution Engine=GPT-4o-mini2026.01 | 89.7 |