List your company

Question Answering on CWQA (test)

45.74F1 ScoreTrustMargin
25.220830.547935.87541.2021Jun 7, 2026

Evaluation results

MethodLinks
TrustMarginBackbone=LLaMA3.1-8B2026.0645.7436.1
FLAREBackbone=LLaMA3.1-8B2026.0643.7434.1
DirectBackbone=LLaMA3.1-8B2026.0643.5133.7
DTR-RAGBackbone=LLaMA3.1-8B2026.0641.7233.7
TrustMarginBackbone=LLaMA3.2-3B2026.0640.7331.6
CLeHe-RAGBackbone=LLaMA3.1-8B2026.0640.6333.2
DirectBackbone=LLaMA3.2-3B2026.0638.8330.6
FLAREBackbone=LLaMA3.2-3B2026.0638.2730.2
BM25-RAGBackbone=LLaMA3.1-8B2026.0638.2530.2
IRCoTBackbone=LLaMA3.1-8B2026.0638.1729.7
CLeHe-RAGBackbone=LLaMA3.2-3B2026.0636.3328
DTR-RAGBackbone=LLaMA3.2-3B2026.0636.3129.3
TrustMarginBackbone=LLaMA3.2-1B2026.0630.623.2
DTR-RAGBackbone=LLaMA3.2-1B2026.0630.5323.7
CLeHe-RAGBackbone=LLaMA3.2-1B2026.0630.4923.2
BM25-RAGBackbone=LLaMA3.2-3B2026.0630.1423
IRCoTBackbone=LLaMA3.2-3B2026.0630.0323.9
FLAREBackbone=LLaMA3.2-1B2026.0628.7421.5
IRCoTBackbone=LLaMA3.2-1B2026.0627.9121
DirectBackbone=LLaMA3.2-1B2026.0627.4621.1
BM25-RAGBackbone=LLaMA3.2-1B2026.0626.0119.7