List your company

Uncertainty Estimation on PopQA

83.09A1EigV+MCH
51.21459.489567.76576.0405May 12, 2025

Evaluation results

MethodLinks
EigV+MCHBase Model=Llama2-chat, Retriever=Contriver, Calibration=MCH2025.0583.0924.6696.7532.8272.18
EigV+CTIBase Model=Llama2-chat, Retriever=Contriver, Calibration=CTI2025.0580.819.1893.529.7761.51
ECC+MCHBase Model=Llama2-chat, Retriever=Contriver, Calibration=MCH2025.0579.3721.9294.8435.8771.39
ECC+CTIBase Model=Llama2-chat, Retriever=Contriver, Calibration=CTI2025.0578.823.2990.8234.3561.75
SE+CTIBase Model=Llama2-chat, Retriever=Contriver, Calibration=CTI2025.0578.5126.0391.2126.7262.58
SE+MCHBase Model=Llama2-chat, Retriever=Contriver, Calibration=MCH2025.0577.9431.5194.0741.2272.49
EigV+NLIBase Model=Llama2-chat, Retriever=Contriver, Calibration=NLI2025.0573.3535.6295.638.1767.6
SEBase Model=Llama2-chat, Retriever=Contriver, Calibration=None2025.0571.9231.5194.0729.0163.79
SE+NLIBase Model=Llama2-chat, Retriever=Contriver, Calibration=NLI2025.0571.6338.3692.7341.2267.86
ECC+NLIBase Model=Llama2-chat, Retriever=Contriver, Calibration=NLI2025.0571.3532.8892.1642.7566.44
PE+CTIBase Model=Llama2-chat, Retriever=Contriver, Calibration=CTI2025.0569.6339.7387.9538.1763.04
PE+MCHBase Model=Llama2-chat, Retriever=Contriver, Calibration=MCH2025.0569.3439.7389.4839.764.31
PE+NLIBase Model=Llama2-chat, Retriever=Contriver, Calibration=NLI2025.0568.7741.188.1541.2263.09
PEBase Model=Llama2-chat, Retriever=Contriver, Calibration=None2025.0568.1842.4687.5738.1761.59
EigVBase Model=Llama2-chat, Retriever=Contriver, Calibration=None2025.0555.36.8592.9320.6162.42
ECCBase Model=Llama2-chat, Retriever=Contriver, Calibration=None2025.0552.446.8487.3818.3262.06