List your company
ResearchFormerly WizWand

Multilabel Classification on Therapist Q&A (test)

0.348Kappa (κF1)MentalRoBERTa (RLP Labels)
0.100480.164740.2290.29326Nov 3, 2025
Updated 8mo ago

Evaluation results

MethodLinks
MentalRoBERTa (RLP Labels)Backbone Model=MentalRoBERTa, Labeling Strategy=RLP Labels, Annotator Model=GPT-4, Confidence Threshold=0.52025.110.348
MentalRoBERTa (RLP Labels)Backbone Model=MentalRoBERTa, Labeling Strategy=RLP Labels, Annotator Model=GPT-4, Confidence Threshold=0.72025.110.331
MentalRoBERTa (MLP Labels)Backbone Model=MentalRoBERTa, Labeling Strategy=MLP Labels, Annotator Model=GPT-4, Confidence Threshold=0.72025.110.324
MentalRoBERTa (MLP Labels)Backbone Model=MentalRoBERTa, Labeling Strategy=MLP Labels, Annotator Model=GPT-4o, Confidence Threshold=0.52025.110.317
MentalRoBERTa (MLP Labels)Backbone Model=MentalRoBERTa, Labeling Strategy=MLP Labels, Annotator Model=GPT-4o, Confidence Threshold=0.72025.110.31
MentalRoBERTa (MLP Labels)Backbone Model=MentalRoBERTa, Labeling Strategy=MLP Labels, Annotator Model=GPT-4, Confidence Threshold=0.52025.110.283
MentalRoBERTa (RLP Labels)Backbone Model=MentalRoBERTa, Labeling Strategy=RLP Labels, Annotator Model=GPT-4o, Confidence Threshold=0.72025.110.279
MentalRoBERTa (RLP Labels)Backbone Model=MentalRoBERTa, Labeling Strategy=RLP Labels, Annotator Model=GPT-4o, Confidence Threshold=0.52025.110.229
MentalRoBERTa (Golden Labels)Backbone Model=MentalRoBERTa, Labeling Strategy=Golden Labels, Annotator Model=GPT-4o, Confidence Threshold=0.72025.110.169
MentalRoBERTa (Golden Labels)Backbone Model=MentalRoBERTa, Labeling Strategy=Golden Labels, Annotator Model=GPT-4, Confidence Threshold=0.72025.110.163
MentalRoBERTa (Golden Labels)Backbone Model=MentalRoBERTa, Labeling Strategy=Golden Labels, Annotator Model=GPT-4, Confidence Threshold=0.52025.110.138
MentalRoBERTa (Golden Labels)Backbone Model=MentalRoBERTa, Labeling Strategy=Golden Labels, Annotator Model=GPT-4o, Confidence Threshold=0.52025.110.11