让对话推荐更懂用户，通过分层决策提升推荐质量。
HARPO: Hierarchical Agentic Reasoning for User-Aligned Conversational Recommendation
- 将推荐质量拆解为相关性、多样性等维度，动态加权
- 用树搜索策略优选候选推荐路径，提升整体质量
- 适合追求真实用户体验的推荐系统研究者
对话式推荐系统在用户偏好逐步揭示的场景下运行，需在不确定性中做出推荐决策。尽管基于大模型的方法在召回率@K和BLEU等代理指标上表现优异，但往往无法实际生成高质量、与用户对齐的推荐，因其优化的是检索准确率或生成流畅性等中间目标，而非推荐质量本身。我们提出HARPO（分层代理推理与偏好优化），一个将对话推荐重构为结构化决策过程的代理框架，以多维推荐质量为目标进行优化。HARPO整合了：(i) 分层偏好学习，将推荐质量分解为可解释的维度（相关性、多样性、满意度、参与度），并依上下文动态加权；(ii) 由学习到的价值网络引导的审议式树搜索，评估候选路径的预测质量；(iii) 通过虚拟工具操作和多代理精炼实现领域无关的推理抽象。我们在ReDial、INSPIRED和MUSE数据集上评估，结果表明HARPO在以推荐为核心的指标上持续优于强基线，同时保持竞争性的回复质量。
原文摘要 · Abstract (English)
Conversational recommender systems (CRSs) operate under incremental preference revelation, requiring recommendation decisions under uncertainty. While recent LLM-based approaches achieve strong performance on proxy metrics such as Recall@K and BLEU, they often fail to deliver high-quality, user-aligned recommendations in practice, as they optimize intermediate objectives like retrieval accuracy or fluent generation rather than recommendation quality itself. We propose HARPO (Hierarchical Agentic Reasoning with Preference Optimization), an agentic framework that reframes conversational recommendation as a structured decision-making process optimized for multi-dimensional recommendation quality. HARPO integrates (i) hierarchical preference learning that decomposes recommendation quality into interpretable dimensions (relevance, diversity, satisfaction, and engagement) with context-dependent weighting; (ii) deliberative tree-search reasoning guided by a learned value network evaluating candidate paths on predicted quality; and (iii) domain-agnostic reasoning abstractions through Virtual Tool Operations and multi-agent refinement. We evaluate HARPO on ReDial, INSPIRED, and MUSE, demonstrating consistent improvements over strong baselines on recommendation-centric metrics while maintaining competitive response quality.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。