List your company

Question Answering on NQ (test) (Unreliability Rate)

59.1Unreliability RateKbSD
57.861266.223174.58582.9469Jun 29, 2026

Evaluation results

MethodLinks
KbSDBackbone=Qwen2.5-7B, Paradigm=RL-based Agentic RAG2026.0659.1
KbPOBackbone=Qwen2.5-7B, Paradigm=RL-based Agentic RAG2026.0662.3
KbSDBackbone=Qwen2.5-3B, Paradigm=RL-based Agentic RAG2026.0664.4
Search-R1Backbone=Qwen2.5-3B, Paradigm=RL-based Agentic RAG2026.0665.27
ReSearchBackbone=Qwen2.5-3B, Paradigm=RL-based Agentic RAG2026.0665.39
ARPOBackbone=Qwen2.5-3B, Paradigm=RL-based Agentic RAG2026.0666.71
KbPOBackbone=Qwen2.5-3B, Paradigm=RL-based Agentic RAG2026.0668.5
ReACTBackbone=Qwen2.5-7B, Paradigm=Agentic RAG2026.0669.99
Search-R1Backbone=Qwen2.5-7B, Paradigm=RL-based Agentic RAG2026.0670.41
TCRAGBackbone=Qwen2.5-7B, Paradigm=Agentic RAG2026.0670.99
IRCOTBackbone=Qwen2.5-3B, Paradigm=Agentic RAG2026.0672.14
ReACTBackbone=Qwen2.5-3B, Paradigm=Agentic RAG2026.0672.81
AEPOBackbone=Qwen2.5-3B, Paradigm=RL-based Agentic RAG2026.0673.24
Mem1Backbone=Qwen2.5-7B, Paradigm=RL-based Agentic RAG2026.0673.62
ReSearchBackbone=Qwen2.5-7B, Paradigm=RL-based Agentic RAG2026.0676.31
COTBackbone=Qwen2.5-7B, Paradigm=No RAG2026.0677.53
FL-RAGBackbone=Qwen2.5-3B, Paradigm=Naive RAG2026.0678.07
TCRAGBackbone=Qwen2.5-3B, Paradigm=Agentic RAG2026.0679.26
FS-RAGBackbone=Qwen2.5-3B, Paradigm=Naive RAG2026.0680.16
BaseBackbone=Qwen2.5-7B, Paradigm=No RAG2026.0680.28
FL-RAGBackbone=Qwen2.5-7B, Paradigm=Naive RAG2026.0680.28
IRCOTBackbone=Qwen2.5-7B, Paradigm=Agentic RAG2026.0680.37
Mem1Backbone=Qwen2.5-3B, Paradigm=RL-based Agentic RAG2026.0680.82
ARPOBackbone=Qwen2.5-7B, Paradigm=RL-based Agentic RAG2026.0682.2
FS-RAGBackbone=Qwen2.5-7B, Paradigm=Naive RAG2026.0683.18
COTBackbone=Qwen2.5-3B, Paradigm=No RAG2026.0683.47
BaseBackbone=Qwen2.5-3B, Paradigm=No RAG2026.0685.73
AEPOBackbone=Qwen2.5-7B, Paradigm=RL-based Agentic RAG2026.0690.07