List your company

Long-form Question Answering on AlpacaFact MixP (50% polluted)

81.7VeriScore F1@kMIRAGE
44.374454.064763.75573.4453Jul 6, 2026

Evaluation results

MethodLinks
MIRAGEBackbone=GPT-4o-mini2026.0781.7
No-RAGBackbone=GPT-4o-mini2026.0776.41
MIRAGEBackbone=Qwen3-8B2026.0772.34
SelfRAGBackbone=GPT-4o-mini2026.0770.38
RetRobust-NLIBackbone=GPT-4o-mini2026.0769.97
AstuteRAGBackbone=GPT-4o-mini2026.0768.11
DRAGINBackbone=GPT-4o-mini2026.0767.43
AstuteRAGBackbone=Qwen3-8B2026.0765.73
No-RAGBackbone=Qwen3-8B2026.0765.71
BRIDGEBackbone=GPT-4o-mini2026.0764.98
SelfRAGBackbone=Qwen3-8B2026.0763.57
BRIDGEBackbone=Qwen3-8B2026.0761.42
RobustRAGBackbone=Qwen3-8B2026.0761.12
TrustRAGBackbone=GPT-4o-mini2026.0760.62
FLAREBackbone=GPT-4o-mini2026.0760.53
MADAMRAGBackbone=GPT-4o-mini2026.0759.59
RetRobust-NLIBackbone=Qwen3-8B2026.0759.13
RAGBackbone=GPT-4o-mini2026.0758.92
RARGBackbone=GPT-4o-mini2026.0757.32
DRAGINBackbone=Qwen3-8B2026.0757.1
FLAREBackbone=Qwen3-8B2026.0755.34
TrustRAGBackbone=Qwen3-8B2026.0755.26
RAGBackbone=Qwen3-8B2026.0755.15
RARGBackbone=Qwen3-8B2026.0753.88
MADAMRAGBackbone=Qwen3-8B2026.0753.61
RobustRAGBackbone=GPT-4o-mini2026.0745.81