Chatbot Arena
由加州大学伯克利分校LMSYS团队创建的AI模型评测平台，采用盲测对战机制，通过用户偏好投票生成ELO积分排行榜，是目前最权威的综合性AI评测之一
时间轴 (近 90 天)
在Chatbot Arena等第三方评测中，Claude在创意写作类任务上的用户偏好率与GPT-4不相上下，在某些风格模仿任务中领先
Chatbot Arena was introduced by a research team at UC Berkeley in 2023 and is one of the earliest platforms to implement large-scale LLM battle evaluation
Chatbot Arena works by having a user submit a question, the system randomly selects two anonymous models to generate responses, and the user votes for the better answer without knowing which model produced it
As of 2024, Chatbot Arena has accumulated over one million human votes
全部知识事实 (4)
Chatbot Arena works by having a user submit a question, the system randomly selects two anonymous models to generate responses, and the user votes for the better answer without knowing which model produced it
95%待验证Chatbot Arena was introduced by a research team at UC Berkeley in 2023 and is one of the earliest platforms to implement large-scale LLM battle evaluation
90%待验证As of 2024, Chatbot Arena has accumulated over one million human votes
85%待验证在Chatbot Arena等第三方评测中，Claude在创意写作类任务上的用户偏好率与GPT-4不相上下，在某些风格模仿任务中领先
50%