There has been a surge in the use of large language models (LLM) conversational agents to generate responses based on long-term history from multiple sessions. However, existing long-term open-domain dialogue datasets lack complex, real-world personalization and fail to capture implicit reasoning-where relevant information is embedded in subtle, syntactic, or semantically distant connections rather than explicit statements. In such cases, traditional retrieval methods fail to capture relevant context, and long-context modeling also becomes inefficient due to numerous complicated persona-related details. To address this gap, we introduce ImplexConv, a large-scale long-term dataset with 2,500 examples, each containing approximately 100 conversation sessions, designed to study implicit reasoning in personalized dialogues. Additionally, we propose TaciTree, a novel hierarchical tree framework that structures conversation history into multiple levels of summarization. Instead of brute-force searching all data, TaciTree enables an efficient, level-based retrieval process where models refine their search by progressively selecting relevant details. Our experiments demonstrate that TaciTree significantly improves the ability of LLMs to reason over long-term conversations with implicit contextual dependencies.

翻译：近年来，基于大规模语言模型（LLM）的对话代理在利用多会话长期历史生成回复方面应用激增。然而，现有的长期开放域对话数据集缺乏复杂、真实的个性化特征，且未能捕捉隐式推理——即相关信息并非通过显式陈述，而是蕴含于细微的、句法的或语义距离较远的关联之中。在此类场景下，传统检索方法难以捕获相关上下文，而长上下文建模也因大量复杂的人物相关细节变得低效。为填补这一空白，我们提出了ImplexConv，一个包含2,500个样本的大规模长期对话数据集，每个样本涵盖约100个对话会话，专为研究个性化对话中的隐式推理而设计。此外，我们提出了TaciTree，一种新颖的层次树框架，该框架将对话历史结构化为多级摘要。TaciTree并非暴力搜索所有数据，而是支持一种高效的、基于层级的检索过程，使模型能够通过逐步选择相关细节来优化搜索。实验表明，TaciTree显著提升了LLM在具有隐式上下文依赖的长期对话中进行推理的能力。