List your company

Out-of-domain Generalization on Diplomat, Mutual, Quality, CoQA, and Qasper (test)

70.9ScoreAutoMix
-5.74814.15134.0553.949Oct 19, 2023

Evaluation results

MethodLinks
AutoMixSLM=GPT-3.52023.1070.9
AutoMixSLM=LLama-13b2023.1031.5
AutoMixSLM=Mistral-7b2023.1028.3
FrugalGPTSLM=GPT-3.52023.1014.3
FrugalGPTSLM=Mistral-7b2023.1012.5
HybridLLMSLM=GPT-3.52023.107.6
HybridLLMSLM=Mistral-7b2023.102.4
FrugalGPTSLM=LLama-13b2023.100
HybridLLMSLM=LLama-13b2023.10-2.8