List your company

Therapeutic Quality Judging on Therapeutic quality evaluation dataset

0.989ICC(C,1)TheraJudge
0.111240.339120.5670.79488Jun 29, 2026

Evaluation results

MethodLinks
TheraJudgeDimension=Guidance2026.060.9890.9480.9810.05
TheraJudgeDimension=Informativeness2026.060.9830.9180.9770.077
TheraJudgeDimension=Relevance2026.060.960.90.9290.085
TheraJudgeDimension=Understanding2026.060.960.8820.9460.106
Best of Closed-SourceDimension=Guidance, Backbone=o4-mini2026.060.9480.7440.7860.233
TheraJudgeDimension=Helpfulness2026.060.9450.6940.9340.291
TheraJudgeDimension=Empathy2026.060.9320.7090.9190.27
Qwen-2.5-7B-SFTDimension=Guidance, Evaluation Protocol=SFT2026.060.9290.8520.9110.114
Qwen-2.5-7B-SFTDimension=Helpfulness, Evaluation Protocol=SFT2026.060.9210.820.9260.16
Best of Closed-SourceDimension=Informativeness, Backbone=o4-mini2026.060.9180.6380.9080.34
Best of Closed-SourceDimension=Empathy, Backbone=Claude-3.72026.060.9060.4290.4740.528
Best of Closed-SourceDimension=Helpfulness, Backbone=Claude-3.72026.060.90.7340.7420.258
Qwen-2.5-7B-SFTDimension=Informativeness, Evaluation Protocol=SFT2026.060.8950.8050.8860.178
TheraJudgeDimension=Safety2026.060.8790.5610.8480.901
Qwen-2.5-7B-SFTDimension=Empathy, Evaluation Protocol=SFT2026.060.8740.4590.8660.518
Best of Closed-SourceDimension=Understanding, Backbone=o4-mini2026.060.8710.6360.5920.302
Qwen-2.5-7B-ZSDimension=Understanding, Evaluation Protocol=Zero-Shot2026.060.870.4710.7120.453
Qwen-2.5-7B-SFTDimension=Understanding, Evaluation Protocol=SFT2026.060.8030.6460.7430.216
Qwen-2.5-7B-ZSDimension=Informativeness, Evaluation Protocol=Zero-Shot2026.060.8020.5730.6690.376
Qwen-2.5-7B-ZSDimension=Helpfulness, Evaluation Protocol=Zero-Shot2026.060.7610.3940.740.493
Qwen-2.5-7B-SFTDimension=Relevance, Evaluation Protocol=SFT2026.060.760.6360.6950.175
Best of Closed-SourceDimension=Relevance, Backbone=Claude-3.72026.060.730.3940.7430.594
Best of Closed-SourceDimension=Safety, Backbone=Claude-3.72026.060.6850.3330.5970.628
Qwen-2.5-7B-ZSDimension=Guidance, Evaluation Protocol=Zero-Shot2026.060.650.450.4990.374
Qwen-2.5-7B-ZSDimension=Empathy, Evaluation Protocol=Zero-Shot2026.060.6160.1640.6020.544
Qwen-2.5-7B-SFTDimension=Safety, Evaluation Protocol=SFT2026.060.6050.4290.4510.242
Qwen-2.5-7B-ZSDimension=Relevance, Evaluation Protocol=Zero-Shot2026.060.5190.310.2630.292
Qwen-2.5-7B-ZSDimension=Safety, Evaluation Protocol=Zero-Shot2026.060.1450.0150.0790.427