List your company

Multi-turn consistency evaluation on MT-Consistency (test)

71.34Acc AvgGPT
40.327248.378656.4364.4814Mar 28, 2025

Evaluation results

MethodLinks
GPTSetting=Multi-Turn2025.0371.346.841.69
ClaudeSetting=Multi-Turn2025.0363.074.381.51
QwenSetting=Multi-Turn2025.0360.866.021.64
MistralSetting=Multi-Turn2025.0350.025.281.53
GeminiSetting=Multi-Turn2025.0341.843.881.25
LlaMaSetting=Multi-Turn2025.0341.524.591.45