List your company

Fact Verification on FEVER (Accuracy and Latency)

61.4AccuracyToT
45.17649.38853.657.812Jun 13, 2024

Evaluation results

MethodLinks
ToTBackbone=Mistral-7B2024.0661.43,291.3
TS-SFTBackbone=Mistral-7B2024.0660.241.7
CPOBackbone=Mistral-7B2024.0659.940.6
CoTBackbone=Mistral-7B2024.0657.941.8
CPOBackbone=LLAMA2-7B2024.0653.236.8
CPOBackbone=LLAMA2-13B2024.0649.243.9
ToTBackbone=LLAMA2-13B2024.0648.83,139.8
TS-SFTBackbone=LLAMA2-13B2024.0648.844
CoTBackbone=LLAMA2-13B2024.0648.245.4
TS-SFTBackbone=LLAMA2-7B2024.0647.534
ToTBackbone=LLAMA2-7B2024.0647.32,087.1
CoTBackbone=LLAMA2-7B2024.0645.833.8