List your company

Human-Metric Correlation on HelpSteer2 (In-Distribution)

0.342Kendall's TauAutoMetrics
0.026880.108690.19050.27231Dec 19, 2025

Evaluation results

MethodLinks
AutoMetricsBackbone=Qwen-3-32B2025.120.342
LLM-JudgeBackbone=Qwen-3-32B2025.120.334
Best Existing MetricBackbone=Model Agnostic2025.120.327
AutoMetricsBackbone=GPT-4o-mini2025.120.324
DnA EvalBackbone=Qwen-3-32B2025.120.26
LLM-JudgeBackbone=GPT-4o-mini2025.120.259
DnA EvalBackbone=GPT-4o-mini2025.120.255
MetaMetricsBackbone=Model Agnostic2025.120.204
Finetuned LLMBackbone=Model Agnostic2025.120.039