Yuan4629/WereBench
收藏Hugging Face2025-10-21 更新2025-10-25 收录
下载链接:
https://hf-mirror.com/datasets/Yuan4629/WereBench
下载链接
链接失效反馈官方服务:
资源简介:
WereBench是一个针对狼人杀社交推理场景的语言模型评估基准数据集,重点关注与人类策略对齐的战略推理。
WereBench is a benchmark dataset for evaluating language models in the Werewolf social deduction setting, focusing on human-aligned strategic reasoning.
提供机构:
Yuan4629



