List your company

Open-domain Single-hop QA on TriviaQA (EM)

76.4EMSDGA-Phase
20.65635.12849.664.072May 12, 2026

Evaluation results

MethodLinks
SDGA-PhaseFramework=ZeroSearch, Model=Qwen2.5-7B, Search Depth (Smax)=52026.0576.4
SDGA-AutoFramework=ZeroSearch, Model=Qwen2.5-7B, Search Depth (Smax)=52026.0573.6
SDGA-PhaseFramework=ZeroSearch, Model=Qwen2.5-3B, Search Depth (Smax)=52026.0572.4
SDGA-AutoFramework=ZeroSearch, Model=Qwen2.5-3B, Search Depth (Smax)=52026.0567.3
GRPO-FullFramework=ZeroSearch, Model=Qwen2.5-7B2026.0565.8
GRPO-HalfFramework=ZeroSearch, Model=Qwen2.5-7B2026.0563.2
GRPO-FullFramework=ZeroSearch, Model=Qwen2.5-3B2026.0560.7
SDGA-AutoFramework=Search-R1, Model=LLaMA-3.2-3B-Base, Search Depth (Smax)=52026.0559.2
SDGA-AutoFramework=Search-R1, Model=Qwen2.5-7B, Search Depth (Smax)=52026.0559.2
GRPO-FullFramework=Search-R1, Model=Qwen2.5-7B2026.0558.8
SDGA-PhaseFramework=Search-R1, Model=LLaMA-3.2-3B-Base, Search Depth (Smax)=52026.0558.7
SDGA-PhaseFramework=Search-R1, Model=Qwen2.5-7B, Search Depth (Smax)=52026.0558.6
GRPO-HalfFramework=ZeroSearch, Model=Qwen2.5-3B2026.0558.5
SDGA-AutoFramework=ZeroSearch, Model=LLaMA-3.2-3B-Base, Search Depth (Smax)=52026.0557.4
SDGA-PhaseFramework=ZeroSearch, Model=LLaMA-3.2-3B-Base, Search Depth (Smax)=52026.0557.1
GRPO-FullFramework=Search-R1, Model=Qwen2.5-3B2026.0556.1
GRPO-HalfFramework=Search-R1, Model=Qwen2.5-7B2026.0555.4
GRPO-FullFramework=ZeroSearch, Model=LLaMA-3.2-3B-Base2026.0552.8
SDGA-AutoFramework=Search-R1, Model=Qwen2.5-3B, Search Depth (Smax)=52026.0552.8
GRPO-HalfFramework=Search-R1, Model=Qwen2.5-3B2026.0552.1
SDGA-PhaseFramework=Search-R1, Model=Qwen2.5-3B, Search Depth (Smax)=52026.0552.1
GRPO-FullFramework=Search-R1, Model=LLaMA-3.2-3B-Base2026.0551.2
GRPO-HalfFramework=ZeroSearch, Model=LLaMA-3.2-3B-Base2026.0548.9
GRPO-HalfFramework=Search-R1, Model=LLaMA-3.2-3B-Base2026.0547.4
SDGA-AntiFramework=ZeroSearch, Model=Qwen2.5-7B, Search Depth (Smax)=52026.0537.8
SDGA-AntiFramework=ZeroSearch, Model=Qwen2.5-3B, Search Depth (Smax)=52026.0535.3
SDGA-AntiFramework=Search-R1, Model=Qwen2.5-7B, Search Depth (Smax)=52026.0534.6
SDGA-AntiFramework=Search-R1, Model=Qwen2.5-3B, Search Depth (Smax)=52026.0527.3
SDGA-AntiFramework=ZeroSearch, Model=LLaMA-3.2-3B-Base, Search Depth (Smax)=52026.0524.9
SDGA-AntiFramework=Search-R1, Model=LLaMA-3.2-3B-Base, Search Depth (Smax)=52026.0522.8