lmarena-ai/arena-expert-5k

这个数据集包含了在文本类别中收集的专家投票。每一行代表对两个模型（model_a和model_b）在用户对话中的表现的单一投票，以及完整的对话历史。关键字段包括：每个投票的唯一反馈ID、当前投票的评估顺序、包含模型a、模型b、平局或两者都不好的投票结果、当前评估顺序的完整对话、包括上下文提示和所有之前评估顺序的答案的整个对话、以及对话的职业标签。

This dataset contains expert votes collected in the text-only category. Each row represents a single vote judging two models (model_a and model_b) on a user conversation, along with the full conversation history. Key fields include: the unique feedback ID of each vote/row, the evaluation order of the current vote, the vote result containing either model_a, model_b, tie, or both_bad, the full conversation of the current evaluation order, the entire conversation including context prompts and answers from all previous evaluation orders, and the occupational categories tagged to each conversation.