List your company
ResearchFormerly WizWand

Schwartz Value (3-class) Evaluation on ValEval (Disturb)

82.12AccuracyCLAVE-Llama
57.898464.186770.47576.7633Jul 15, 2024
Updated 8mo ago

Evaluation results

MethodLinks
CLAVE-LlamaEvaluation Protocol=Tuning-based Evaluation, Backbone=Llama2-7b2024.0782.12
FairEvalEvaluation Protocol=Prompt-based Evaluation2024.0781.5
ChatEvalEvaluation Protocol=Prompt-based Evaluation2024.0781.42
WideDeepEvaluation Protocol=Prompt-based Evaluation2024.0780.42
CLAVE-MistralEvaluation Protocol=Tuning-based Evaluation, Backbone=Mistral-7b2024.0775.05
Mistral-7bEvaluation Protocol=Tuning-based Evaluation, Backbone=Mistral-7b2024.0770.89
CrowdworkerEvaluation Protocol=Tuning-based Evaluation2024.0768.65
CalibrateEvaluation Protocol=Prompt-based Evaluation2024.0768.49
Chain-of-thoughtEvaluation Protocol=Prompt-based Evaluation2024.0768.39
Phi-3Evaluation Protocol=Tuning-based Evaluation, Backbone=Phi-32024.0768.19
VanillaEvaluation Protocol=Prompt-based Evaluation2024.0768.13
G-EvalEvaluation Protocol=Prompt-based Evaluation2024.0767.88
Few-shotEvaluation Protocol=Prompt-based Evaluation2024.0767.62
AllureEvaluation Protocol=Prompt-based Evaluation2024.0767.42
GPT-2-LargeEvaluation Protocol=Tuning-based Evaluation, Backbone=GPT-2-Large2024.0760.49
Llama2-7bEvaluation Protocol=Tuning-based Evaluation, Backbone=Llama2-7b2024.0758.83