Chat models, such as ChatGPT, have shown impressive capabilities and have been rapidly adopted across numerous domains. However, these models are only accessible through a restricted API, creating barriers for new research and progress in the field. We propose a pipeline that can automatically generate a high-quality multi-turn chat corpus by leveraging ChatGPT to engage in a conversation with itself. Subsequently, we employ parameter-efficient tuning to enhance LLaMA, an open-source large language model. The resulting model, named Baize, demonstrates good performance in multi-turn dialogues with guardrails that minimize potential risks. Furthermore, we propose a new technique called Self-Distill with Feedback, to further improve the performance of the Baize models with feedback from ChatGPT. The Baize models and data are released for research purposes only at https://github.com/project-baize/baize-chatbot. An online demo is also available at https://huggingface.co/spaces/project-baize/chat-with-baize.

翻译：诸如ChatGPT之类的对话模型已展现出令人瞩目的能力，并在众多领域得到快速应用。然而，这些模型仅能通过受限的API访问，给该领域的新研究与进展造成障碍。我们提出了一种流水线方法，通过让ChatGPT进行自我对话，自动生成高质量的多轮对话语料库。随后，我们采用参数高效调优技术对开源大语言模型LLaMA进行增强。由此产生的模型命名为Baize，在具有风险规避护栏的多轮对话中表现出良好性能。此外，我们提出了一种名为"带反馈的自蒸馏"新技术，利用ChatGPT的反馈进一步优化Baize模型的性能。Baize模型及数据仅限研究用途，发布于https://github.com/project-baize/baize-chatbot，同时在线演示地址为https://huggingface.co/spaces/project-baize/chat-with-baize。