Slot Filling on T-REx
65.07Coverage EMSearChain
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| SearChainInformation Retrieval=true2023.04 | 65.07 | — | — | |
| Tree-of-Thought w/ IRInformation Retrieval=true2023.04 | 62.4 | — | — | |
| DSPInformation Retrieval=true2023.04 | 61.32 | — | — | |
| Verify-and-EditInformation Retrieval=true2023.04 | 61.1 | — | — | |
| React -> CoT-SCInformation Retrieval=true2023.04 | 60.42 | — | — | |
| Plan-and-Solve w/ IRInformation Retrieval=true2023.04 | 60.03 | — | — | |
| Direct RetrievalInformation Retrieval=true2023.04 | 59.28 | — | — | |
| ToolFormerInformation Retrieval=true2023.04 | 59.17 | — | — | |
| Self-AskInformation Retrieval=true2023.04 | 59.12 | — | — | |
| SearChain w/o CompletionInformation Retrieval=true, Completion=false2023.04 | 56.03 | — | — | |
| SearChain w/o VerificationInformation Retrieval=true, Verification=false2023.04 | 55.46 | — | — | |
| SearChain w/o IRInformation Retrieval=false2023.04 | 52.69 | — | — | |
| Plan-and-SolveInformation Retrieval=false2023.04 | 47.58 | — | — | |
| Recite-and-answerInformation Retrieval=false2023.04 | 46.14 | — | — | |
| COT-SCInformation Retrieval=false2023.04 | 46.06 | — | — | |
| Least-to-MostInformation Retrieval=false2023.04 | 44.98 | — | — | |
| Self-Ask w/o IRInformation Retrieval=false2023.04 | 44.71 | — | — | |
| CoTInformation Retrieval=false2023.04 | 44.51 | — | — | |
| Auto-CoTInformation Retrieval=false2023.04 | 43.98 | — | — | |
| Direct PromptingInformation Retrieval=false2023.04 | 43.85 | — | — | |
| ARAMBackbone=LLaDA-8B-Instruct2026.03 | — | 31.4 | 34.8 | |
| ARAMBackbone=Dream-v0-Instruct-7B2026.03 | — | 30 | 35.5 | |
| DreamBackbone=Dream-v0-Instruct-7B2026.03 | — | 30.4 | 35.1 | |
| Dream + A-CFGBackbone=Dream-v0-Instruct-7B2026.03 | — | 28.1 | 34.1 | |
| Dream + ADACADBackbone=Dream-v0-Instruct-7B2026.03 | — | 31.4 | 36.7 | |
| Dream + CADBackbone=Dream-v0-Instruct-7B2026.03 | — | 19.3 | 26.8 | |
| Dream + COIECDBackbone=Dream-v0-Instruct-7B2026.03 | — | 24.4 | 30.6 | |
| Dream + RAGBackbone=Dream-v0-Instruct-7B2026.03 | — | 29.7 | 35 | |
| Dream + SPREADBackbone=Dream-v0-Instruct-7B2026.03 | — | 34.5 | 38 | |
| LLaDABackbone=LLaDA-8B-Instruct2026.03 | — | 23.9 | 26.4 | |
| LLaDA + A-CFGBackbone=LLaDA-8B-Instruct2026.03 | — | 18.2 | 35.9 | |
| LLaDA + ADACADBackbone=LLaDA-8B-Instruct2026.03 | — | 21.7 | 24.7 | |
| LLaDA + CADBackbone=LLaDA-8B-Instruct2026.03 | — | 1.3 | 7.4 | |
| LLaDA + COIECDBackbone=LLaDA-8B-Instruct2026.03 | — | 11.1 | 17.1 | |
| LLaDA + RAGBackbone=LLaDA-8B-Instruct2026.03 | — | 17.8 | 23.8 | |
| LLaDA + SPREADBackbone=LLaDA-8B-Instruct2026.03 | — | 18 | 24 | |
| LLaMABackbone=LLaMA-3.1-8B-Instruct2026.03 | — | 34.2 | 39.6 | |
| LLaMA + RAGBackbone=LLaMA-3.1-8B-Instruct2026.03 | — | 30.3 | 35.3 | |
| QwenBackbone=Qwen2.5-7B-Instruct2026.03 | — | 32.6 | 36.5 | |
| Qwen + RAGBackbone=Qwen2.5-7B-Instruct2026.03 | — | 30.1 | 34.7 |