We propose MemoChat, a pipeline for refining instructions that enables large language models (LLMs) to effectively employ self-composed memos for maintaining consistent long-range open-domain conversations. We demonstrate a long-range open-domain conversation through iterative "memorization-retrieval-response" cycles. This requires us to carefully design tailored tuning instructions for each distinct stage. The instructions are reconstructed from a collection of public datasets to teach the LLMs to memorize and retrieve past dialogues with structured memos, leading to enhanced consistency when participating in future conversations. We invite experts to manually annotate a test set designed to evaluate the consistency of long-range conversations questions. Experiments on three testing scenarios involving both open-source and API-accessible chatbots at scale verify the efficacy of MemoChat, which outperforms strong baselines. Our codes, data and models are available here: https://github.com/LuJunru/MemoChat.

翻译：我们提出了MemoChat，一种用于优化指令的流水线，使大语言模型（LLMs）能够有效利用自编备忘录来维持一致的长程开放域对话。我们通过迭代式“记忆-检索-响应”循环来演示长程开放域对话。这要求我们为每个不同阶段精心设计定制化的调优指令。这些指令基于一系列公开数据集重建，以教导LLMs使用结构化备忘录来记忆和检索过往对话，从而在参与未来对话时增强一致性。我们邀请专家手动标注了一个测试集，旨在评估长程对话的一致性。在涉及开源和API可访问聊天机器人的三种测试场景中，大规模实验验证了MemoChat的有效性，其性能优于强基线模型。我们的代码、数据和模型可在以下链接获取：https://github.com/LuJunru/MemoChat。