Question Answering on BBH Disambiguation QA
71Accuracy (BBH Disambiguation QA)TextGrad
Evaluation Results
| Method | Links | |
|---|---|---|
| TextGradBackbone=GPT-4o, Optimization Type=Template2025.11 | 71 | |
| evaluation-instructed prompt optimizationBackbone=GPT-4o, Optimization Type=Query-dependent2025.11 | 69 | |
| APEBackbone=GPT-4o, Optimization Type=Template2025.11 | 67 | |
| Self-RefineBackbone=LLaMA-3.1, Optimization Type=Query-dependent2025.11 | 66 | |
| evaluation-instructed prompt optimizationBackbone=LLaMA-3, Optimization Type=Query-dependent2025.11 | 65 | |
| Self-RefineBackbone=LLaMA-3, Optimization Type=Query-dependent2025.11 | 65 | |
| evaluation-instructed prompt optimizationBackbone=LLaMA-3.1, Optimization Type=Query-dependent2025.11 | 65 | |
| TextGradBackbone=LLaMA-3.1, Optimization Type=Template2025.11 | 65 | |
| TextGradBackbone=LLaMA-3, Optimization Type=Template2025.11 | 64 | |
| LLM onlyBackbone=LLaMA-3.1, Optimization Type=None2025.11 | 64 | |
| Pro-RefineBackbone=LLaMA-3.1, Optimization Type=Query-dependent2025.11 | 64 | |
| LLM onlyBackbone=LLaMA-3, Optimization Type=None2025.11 | 63 | |
| Pro-RefineBackbone=LLaMA-3, Optimization Type=Query-dependent2025.11 | 63 | |
| APEBackbone=LLaMA-3, Optimization Type=Template2025.11 | 63 | |
| APEBackbone=LLaMA-3.1, Optimization Type=Template2025.11 | 63 | |
| Pro-RefineBackbone=GPT-4o, Optimization Type=Query-dependent2025.11 | 63 | |
| LLM onlyBackbone=GPT-4o, Optimization Type=None2025.11 | 58 | |
| Self-RefineBackbone=GPT-4o, Optimization Type=Query-dependent2025.11 | 56 |