List your company

Question Answering on SimpleQA out-domain (test)

36.5LasJD2PLAN-7B
16.94822.02427.132.176Jan 13, 2026

Evaluation results

MethodLinks
D2PLAN-7BBase LLM=Qwen2.5-7B-Instruct2026.0136.527.5—
AutoRefineBase LLM=Qwen2.5-3B-Instruct2026.0136.224.9—
ReSearchBase LLM=Qwen2.5-7B-Instruct2026.0136.227.3—
StepSearchBase LLM=Qwen2.5-7B-Instruct2026.0134.824.9—
R1-SearcherBase LLM=Qwen2.5-7B-Instruct2026.0134.625.6—
Search-R1Base LLM=Qwen2.5-7B-Instruct2026.0134.423.1—
D2PLAN-3BBase LLM=Qwen2.5-3B-Instruct2026.0134.125.5—
ZeroSearchBase LLM=Qwen2.5-7B-Instruct2026.0133.124.1—
ZeroSearchBase LLM=Qwen2.5-3B-Instruct2026.0131.522.8—
StepSearchBase LLM=Qwen2.5-3B-Instruct2026.0128.220.4—
Search-R1Base LLM=Qwen2.5-3B-Instruct2026.0117.78.2—
A-Mem2026.04——38.8
ExpeL2026.04——43
Mem02026.04——38.1
Memento2026.04——42.4
MIAvariant=Unsupervised2026.04——46.6
MIAvariant=Standard2026.04——47.7
No Memory2026.04——40.7
RAG2026.04——38.3
ReasoningBank2026.04——42.4