List your company
ResearchFormerly WizWand

Moral Foundation (3-class) Evaluation on ValEval Generalized

53.84AccuracyCLAVE-Llama
12.468823.209433.9544.6906Jul 15, 2024
Updated 8mo ago

Evaluation results

MethodLinks
CLAVE-LlamaEvaluation Protocol=Tuning-based Evaluation, Backbone=Llama2-7b2024.0753.84
Llama2-7bEvaluation Protocol=Tuning-based Evaluation, Backbone=Llama2-7b2024.0751.76
GPT-2-LargeEvaluation Protocol=Tuning-based Evaluation, Backbone=GPT-2-Large2024.0751.14
CLAVE-MistralEvaluation Protocol=Tuning-based Evaluation, Backbone=Mistral-7b2024.0749.27
CrowdworkerEvaluation Protocol=Tuning-based Evaluation2024.0749.25
Phi-3Evaluation Protocol=Tuning-based Evaluation, Backbone=Phi-32024.0749.06
Mistral-7bEvaluation Protocol=Tuning-based Evaluation, Backbone=Mistral-7b2024.0748.02
WideDeepEvaluation Protocol=Prompt-based Evaluation2024.0747.92
ChatEvalEvaluation Protocol=Prompt-based Evaluation2024.0747.5
FairEvalEvaluation Protocol=Prompt-based Evaluation2024.0746.67
AllureEvaluation Protocol=Prompt-based Evaluation2024.0746.09
Chain-of-thoughtEvaluation Protocol=Prompt-based Evaluation2024.0742.97
G-EvalEvaluation Protocol=Prompt-based Evaluation2024.0742.19
VanillaEvaluation Protocol=Prompt-based Evaluation2024.0739.84
CalibrateEvaluation Protocol=Prompt-based Evaluation2024.0738.28
Few-shotEvaluation Protocol=Prompt-based Evaluation2024.0714.06