OnePred: 多轮对话中基于递归意图记忆的下一个查询预测
OnePred: Next-Query Prediction via Recursive Intent Memory in Multi-Turn Conversations
AI总结：
针对多轮对话中下一个查询预测任务，提出OnePred模型，通过递归更新的意图记忆代替完整历史，结合两阶段强化学习训练，在NQP-Bench基准上实现高达22倍的每轮token节省且预测质量优于所有基线。
摘要：
尽管大语言模型（LLM）对话系统每天处理数百万次多轮对话，但它们本质上仍是被动的：仅在用户输入查询后才响应。迈向主动交互的关键一步是下一个查询预测，即仅根据之前的对话预测用户后续的查询。该任务的进展受到缺乏专用基准以及基本效率-质量权衡的阻碍：简单拼接完整对话历史会导致线性增长的token消耗，而截断至最新一轮则会丢弃关键的跨轮上下文。我们的关键见解是，准确预测不需要重新阅读原始历史；只需跟踪用户跨主题、未解决需求和兴趣转移的不断演变的意图轨迹即可。我们提出OnePred，它维护一个递归更新的记忆作为唯一的跨轮上下文，将每轮成本限制为与对话长度无关。我们通过两阶段强化学习流程训练模型，首先教导预测什么，然后教导压缩什么，将记忆塑造成面向预测的意图链。为了建立严格的测试平台，我们引入了NQP-Bench，涵盖三个不同的子集。实验表明，与完整历史输入相比，OnePred将每轮token消耗减少高达22倍，同时在预测质量上持续超过所有基线，在较长对话中增益更大。我们的代码可在https://github.com/ZBWpro/OnePred公开获取。
英文摘要：
Although large language model (LLM) conversational systems process millions of multi-turn dialogues daily, they remain fundamentally reactive: they respond only after the user types a query. A key step toward proactive interaction is next-query prediction, which anticipates the user's subsequent query based solely on the preceding dialogue. Progress on this task is hindered by the lack of dedicated benchmarks and a fundamental efficiency--quality trade-off: naively concatenating full dialogue history incurs linearly growing token consumption, while truncating to the latest turn discards crucial cross-turn context. Our key insight is that accurate prediction does not require re-reading raw history; it suffices to track the user's evolving intent trajectory across topics, unresolved needs, and interest shifts. We propose OnePred, which maintains a recursively updated memory as its sole cross-turn context, bounding the per-turn cost independently of conversation length. We train the model via a two-stage reinforcement learning pipeline that first teaches what to predict, then what to compress, shaping the memory into a prediction-oriented intent chain. To establish a rigorous testbed, we introduce NQP-Bench, spanning three diverse subsets. Experiments demonstrate that OnePred reduces per-turn token consumption by up to 22$\times$ compared to full-history inputs while consistently exceeding all baselines in prediction quality, with larger gains on longer conversations. Our code is publicly available at https://github.com/ZBWpro/OnePred.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
