List your company

Domain-specific reasoning on ClimateQA

79.97Accuracy (ACC)UQ4CT
68.186871.245974.30577.3641Oct 9, 2024

Evaluation results

MethodLinks
UQ4CTBackbone=Llama3.1-8B2024.1079.979.362
MixLoRABackbone=Llama3.1-8B2024.1079.94114.68
LABackbone=Llama3.1-8B2024.1079.82311.481
BLoB(Mean)Backbone=Llama3.1-8B, Variant=Mean2024.1079.0212.74
BLoB(N=10)Backbone=Llama3.1-8B, N=102024.1079.0211.962
EnsembleBackbone=Llama3.1-8B2024.1078.53214.69
LoRABackbone=Llama3.1-8B2024.1078.25113.701
MC DropBackbone=Llama3.1-8B2024.1078.19113.111
Base ModelBackbone=Llama3.1-8B2024.1068.6414.58