Open-domain Single-hop QA on TriviaQA (EM)
76.4EMSDGA-Phase
Evaluation Results
| Method | Links | |
|---|---|---|
| SDGA-PhaseFramework=ZeroSearch, Model=Qwen2.5-7B, Search Depth (Smax)=52026.05 | 76.4 | |
| SDGA-AutoFramework=ZeroSearch, Model=Qwen2.5-7B, Search Depth (Smax)=52026.05 | 73.6 | |
| SDGA-PhaseFramework=ZeroSearch, Model=Qwen2.5-3B, Search Depth (Smax)=52026.05 | 72.4 | |
| SDGA-AutoFramework=ZeroSearch, Model=Qwen2.5-3B, Search Depth (Smax)=52026.05 | 67.3 | |
| GRPO-FullFramework=ZeroSearch, Model=Qwen2.5-7B2026.05 | 65.8 | |
| GRPO-HalfFramework=ZeroSearch, Model=Qwen2.5-7B2026.05 | 63.2 | |
| GRPO-FullFramework=ZeroSearch, Model=Qwen2.5-3B2026.05 | 60.7 | |
| SDGA-AutoFramework=Search-R1, Model=LLaMA-3.2-3B-Base, Search Depth (Smax)=52026.05 | 59.2 | |
| SDGA-AutoFramework=Search-R1, Model=Qwen2.5-7B, Search Depth (Smax)=52026.05 | 59.2 | |
| GRPO-FullFramework=Search-R1, Model=Qwen2.5-7B2026.05 | 58.8 | |
| SDGA-PhaseFramework=Search-R1, Model=LLaMA-3.2-3B-Base, Search Depth (Smax)=52026.05 | 58.7 | |
| SDGA-PhaseFramework=Search-R1, Model=Qwen2.5-7B, Search Depth (Smax)=52026.05 | 58.6 | |
| GRPO-HalfFramework=ZeroSearch, Model=Qwen2.5-3B2026.05 | 58.5 | |
| SDGA-AutoFramework=ZeroSearch, Model=LLaMA-3.2-3B-Base, Search Depth (Smax)=52026.05 | 57.4 | |
| SDGA-PhaseFramework=ZeroSearch, Model=LLaMA-3.2-3B-Base, Search Depth (Smax)=52026.05 | 57.1 | |
| GRPO-FullFramework=Search-R1, Model=Qwen2.5-3B2026.05 | 56.1 | |
| GRPO-HalfFramework=Search-R1, Model=Qwen2.5-7B2026.05 | 55.4 | |
| GRPO-FullFramework=ZeroSearch, Model=LLaMA-3.2-3B-Base2026.05 | 52.8 | |
| SDGA-AutoFramework=Search-R1, Model=Qwen2.5-3B, Search Depth (Smax)=52026.05 | 52.8 | |
| GRPO-HalfFramework=Search-R1, Model=Qwen2.5-3B2026.05 | 52.1 | |
| SDGA-PhaseFramework=Search-R1, Model=Qwen2.5-3B, Search Depth (Smax)=52026.05 | 52.1 | |
| GRPO-FullFramework=Search-R1, Model=LLaMA-3.2-3B-Base2026.05 | 51.2 | |
| GRPO-HalfFramework=ZeroSearch, Model=LLaMA-3.2-3B-Base2026.05 | 48.9 | |
| GRPO-HalfFramework=Search-R1, Model=LLaMA-3.2-3B-Base2026.05 | 47.4 | |
| SDGA-AntiFramework=ZeroSearch, Model=Qwen2.5-7B, Search Depth (Smax)=52026.05 | 37.8 | |
| SDGA-AntiFramework=ZeroSearch, Model=Qwen2.5-3B, Search Depth (Smax)=52026.05 | 35.3 | |
| SDGA-AntiFramework=Search-R1, Model=Qwen2.5-7B, Search Depth (Smax)=52026.05 | 34.6 | |
| SDGA-AntiFramework=Search-R1, Model=Qwen2.5-3B, Search Depth (Smax)=52026.05 | 27.3 | |
| SDGA-AntiFramework=ZeroSearch, Model=LLaMA-3.2-3B-Base, Search Depth (Smax)=52026.05 | 24.9 | |
| SDGA-AntiFramework=Search-R1, Model=LLaMA-3.2-3B-Base, Search Depth (Smax)=52026.05 | 22.8 |