List your company

Question Answering on FRAMES (test)

41.1Unreliability RateKbSD
38.943653.499368.05582.6107Jun 29, 2026

Evaluation results

MethodLinks
KbSDBackbone=Qwen2.5-7B, Paradigm=RL-based Agentic RAG2026.0641.1
KbSDBackbone=Qwen2.5-3B, Paradigm=RL-based Agentic RAG2026.0644
KbPOBackbone=Qwen2.5-7B, Paradigm=RL-based Agentic RAG2026.0656.3
KbPOBackbone=Qwen2.5-3B, Paradigm=RL-based Agentic RAG2026.0662.6
Search-R1Backbone=Qwen2.5-7B, Paradigm=RL-based Agentic RAG2026.0681.99
COTBackbone=Qwen2.5-7B, Paradigm=No RAG2026.0682.4
TCRAGBackbone=Qwen2.5-7B, Paradigm=Agentic RAG2026.0683.54
ReSearchBackbone=Qwen2.5-7B, Paradigm=RL-based Agentic RAG2026.0684.39
ReACTBackbone=Qwen2.5-7B, Paradigm=Agentic RAG2026.0684.71
Mem1Backbone=Qwen2.5-7B, Paradigm=RL-based Agentic RAG2026.0685.85
ARPOBackbone=Qwen2.5-3B, Paradigm=RL-based Agentic RAG2026.0686.51
BaseBackbone=Qwen2.5-7B, Paradigm=No RAG2026.0687.48
FS-RAGBackbone=Qwen2.5-7B, Paradigm=Naive RAG2026.0687.48
AEPOBackbone=Qwen2.5-3B, Paradigm=RL-based Agentic RAG2026.0687.74
ARPOBackbone=Qwen2.5-7B, Paradigm=RL-based Agentic RAG2026.0687.82
FL-RAGBackbone=Qwen2.5-7B, Paradigm=Naive RAG2026.0687.9
Search-R1Backbone=Qwen2.5-3B, Paradigm=RL-based Agentic RAG2026.0689.24
ReACTBackbone=Qwen2.5-3B, Paradigm=Agentic RAG2026.0689.47
FS-RAGBackbone=Qwen2.5-3B, Paradigm=Naive RAG2026.0689.58
ReSearchBackbone=Qwen2.5-3B, Paradigm=RL-based Agentic RAG2026.0690
FL-RAGBackbone=Qwen2.5-3B, Paradigm=Naive RAG2026.0690.81
BaseBackbone=Qwen2.5-3B, Paradigm=No RAG2026.0691.99
TCRAGBackbone=Qwen2.5-3B, Paradigm=Agentic RAG2026.0692.31
AEPOBackbone=Qwen2.5-7B, Paradigm=RL-based Agentic RAG2026.0692.76
COTBackbone=Qwen2.5-3B, Paradigm=No RAG2026.0692.84
IRCOTBackbone=Qwen2.5-3B, Paradigm=Agentic RAG2026.0693.21
IRCOTBackbone=Qwen2.5-7B, Paradigm=Agentic RAG2026.0693.22
Mem1Backbone=Qwen2.5-3B, Paradigm=RL-based Agentic RAG2026.0695.01