List your company

Confidence Calibration in Retrieval-Augmented Generation on Bamboogle k=5 OOD (test)

0.065ECENAACL
0.057560.107780.1580.20822Jan 16, 2026

Evaluation results

MethodLinks
NAACLBackbone=Qwen2.5-7B-Instruct, Number of retrieved passages (k)=52026.010.0650.845
CoTBackbone=Qwen2.5-7B-Instruct, Number of retrieved passages (k)=52026.010.1210.81
VanillaBackbone=Qwen2.5-7B-Instruct, Number of retrieved passages (k)=52026.010.1260.835
NAACLBackbone=Llama-3.1-8B-Instruct, Number of retrieved passages (k)=52026.010.1270.823
EnsembleBackbone=Qwen2.5-7B-Instruct, Number of retrieved passages (k)=52026.010.1270.831
NAACLBackbone=DeepSeek-R1-Distill-Qwen-7B, Number of retrieved passages (k)=52026.010.1270.765
Label-only SFTBackbone=Qwen2.5-7B-Instruct, Number of retrieved passages (k)=52026.010.1290.658
Noise-awareBackbone=Qwen2.5-7B-Instruct, Number of retrieved passages (k)=52026.010.1350.753
NAACLBackbone=DeepSeek-R1-Distill-Llama-8B, Number of retrieved passages (k)=52026.010.1370.697
Noise-awareBackbone=Llama-3.1-8B-Instruct, Number of retrieved passages (k)=52026.010.140.806
EnsembleBackbone=DeepSeek-R1-Distill-Llama-8B, Number of retrieved passages (k)=52026.010.1430.711
Noise-awareBackbone=DeepSeek-R1-Distill-Qwen-7B, Number of retrieved passages (k)=52026.010.1430.625
Label-only SFTBackbone=Llama-3.1-8B-Instruct, Number of retrieved passages (k)=52026.010.1510.721
Noise-awareBackbone=DeepSeek-R1-Distill-Llama-8B, Number of retrieved passages (k)=52026.010.1530.658
VanillaBackbone=DeepSeek-R1-Distill-Llama-8B, Number of retrieved passages (k)=52026.010.1540.671
EnsembleBackbone=Llama-3.1-8B-Instruct, Number of retrieved passages (k)=52026.010.1730.68
EnsembleBackbone=DeepSeek-R1-Distill-Qwen-7B, Number of retrieved passages (k)=52026.010.1980.55
CoTBackbone=Llama-3.1-8B-Instruct, Number of retrieved passages (k)=52026.010.1990.579
CoTBackbone=DeepSeek-R1-Distill-Llama-8B, Number of retrieved passages (k)=52026.010.2030.633
VanillaBackbone=Llama-3.1-8B-Instruct, Number of retrieved passages (k)=52026.010.2120.633
Label-only SFTBackbone=DeepSeek-R1-Distill-Qwen-7B, Number of retrieved passages (k)=52026.010.2160.65
CoTBackbone=DeepSeek-R1-Distill-Qwen-7B, Number of retrieved passages (k)=52026.010.2180.638
VanillaBackbone=DeepSeek-R1-Distill-Qwen-7B, Number of retrieved passages (k)=52026.010.220.626
Label-only SFTBackbone=DeepSeek-R1-Distill-Llama-8B, Number of retrieved passages (k)=52026.010.2510.684