List your company

Long-form Question Answering on AlpacaFact FullP (100% polluted)

78.65VeriScore F1@kMIRAGE
24.372438.463752.55566.6463Jul 6, 2026

Evaluation results

MethodLinks
MIRAGEBackbone=GPT-4o-mini2026.0778.65
No-RAGBackbone=GPT-4o-mini2026.0776.41
MIRAGEBackbone=Qwen3-8B2026.0767.79
No-RAGBackbone=Qwen3-8B2026.0765.71
SelfRAGBackbone=GPT-4o-mini2026.0765.68
RetRobust-NLIBackbone=GPT-4o-mini2026.0764.53
AstuteRAGBackbone=GPT-4o-mini2026.0763.76
BRIDGEBackbone=GPT-4o-mini2026.0762.38
DRAGINBackbone=GPT-4o-mini2026.0761.47
AstuteRAGBackbone=Qwen3-8B2026.0760.3
RetRobust-NLIBackbone=Qwen3-8B2026.0757.2
BRIDGEBackbone=Qwen3-8B2026.0756.66
FLAREBackbone=GPT-4o-mini2026.0755.67
RAGBackbone=GPT-4o-mini2026.0752.9
SelfRAGBackbone=Qwen3-8B2026.0750.63
FLAREBackbone=Qwen3-8B2026.0746.17
RARGBackbone=GPT-4o-mini2026.0745.63
RobustRAGBackbone=Qwen3-8B2026.0744.32
DRAGINBackbone=Qwen3-8B2026.0743.93
TrustRAGBackbone=GPT-4o-mini2026.0743.15
MADAMRAGBackbone=GPT-4o-mini2026.0740.54
RAGBackbone=Qwen3-8B2026.0738.43
TrustRAGBackbone=Qwen3-8B2026.0737.69
MADAMRAGBackbone=Qwen3-8B2026.0736.04
RARGBackbone=Qwen3-8B2026.0735.87
RobustRAGBackbone=GPT-4o-mini2026.0726.46