Multi-hop Question Answering on StrategyQA (SQA)
76.95Cover-EMSearChain
Evaluation results
| Method | Links | ||
|---|---|---|---|
| SearChainInformation Retrieval=true2023.04 | 76.95 | — | |
| SearChain w/o CompletionInformation Retrieval=true, Completion=false2023.04 | 76.53 | — | |
| SearChain w/o VerificationInformation Retrieval=true, Verification=false2023.04 | 75.98 | — | |
| SearChain w/o IRInformation Retrieval=false2023.04 | 75.62 | — | |
| Plan-and-Solve w/ IRInformation Retrieval=true2023.04 | 74.58 | — | |
| Plan-and-SolveInformation Retrieval=false2023.04 | 73.21 | — | |
| Tree-of-Thought w/ IRInformation Retrieval=true2023.04 | 72.55 | — | |
| DSPInformation Retrieval=true2023.04 | 72.41 | — | |
| Verify-and-EditInformation Retrieval=true2023.04 | 71.09 | — | |
| COT-SCInformation Retrieval=false2023.04 | 70.84 | — | |
| Recite-and-answerInformation Retrieval=false2023.04 | 70.47 | — | |
| React -> CoT-SCInformation Retrieval=true2023.04 | 70.43 | — | |
| Self-AskInformation Retrieval=true2023.04 | 67.65 | — | |
| ToolFormerInformation Retrieval=true2023.04 | 67.02 | — | |
| Direct RetrievalInformation Retrieval=true2023.04 | 66.78 | — | |
| Direct PromptingInformation Retrieval=false2023.04 | 66.25 | — | |
| CoTInformation Retrieval=false2023.04 | 65.83 | — | |
| Least-to-MostInformation Retrieval=false2023.04 | 65.78 | — | |
| Self-Ask w/o IRInformation Retrieval=false2023.04 | 65.45 | — | |
| Auto-CoTInformation Retrieval=false2023.04 | 65.4 | — | |
| IRCoT2026.05 | — | 70.9 | |
| Search-R12026.05 | — | 72.3 | |
| Static RAG2026.05 | — | 69.5 | |
| Step-Levelretrieval_strategy=step-level adaptive retrieval2026.05 | — | 73.4 |