List your company

Multi-party cognitive stimulation dialogue on Real-world Cognitive Stimulation Dialogue dataset (test)

75Win RateGCSD-3b
41.7250.365967.64Mar 1, 2026

Evaluation results

MethodLinks
GCSD-3bbaseline_comparison=ERNIE2026.03751015
GCSD-3bbaseline_comparison=GLM-4-Plus2026.03731215
GCSD-3bbaseline_comparison=Doubao-Pro2026.03602317
GCSD-3bbaseline_comparison=ChatGPT2026.03582715
GCSD-3bbaseline_comparison=LLaMA-3.1-405B2026.0358348
GCSD-3bbaseline_comparison=Qwen-Max2026.03552916
GCSD-3bbaseline_comparison=GPT-4o2026.03503911
GCSD-3bbaseline_comparison=DeepSeek-671B2026.03434017