[论文] HARP: Training-Free Dual-Profile Agentic Communication for LLM-Based R...

论文概要

研究领域: ML 作者: Zhen Tao, Xun Zhou, Xuhui Chen, Ziyue Qiao, Qingqiang Sun 发布时间: 2026-11-03 arXiv: 2611.05319

中文摘要

在基于 LLM 的推荐中，记忆充当记忆增强骨干上的用户级智能体。但用户与物品是不对称的：只有用户被赋予结构化画像；它进入一个带宽受限的重排提示；且画像在预热后被冻结。我们把基于 LLM 的推荐重构为一个四阶段的智能体沟通过程：BUILD（生成自我描述）、TRANSMIT（沿协同图传播）、UNDERSTAND（理解对方）、SYNC（从下一次交互中同步）。这一形式化揭示三个未解决的沟通挑战：物品侧从未产生可比较的自我描述（单边 BUILD）；画像级信号只进入带宽受限的重排提示而非检索，在检索中它们本可不占上下文 token（被困于提示的 UNDERSTAND）；用户描述在预热后被冻结，缺乏有原则的更新规则（缺席的 SYNC）。我们把缺失的阶段实例化为 HARP——一种免训练、三个算子的附加组件，叠加在冻结的 LLM 骨干上：规则抽取的对称物品画像（零额外 LLM 调用）、信道可分解的画像级同质性 H(φ_u, φ_i)（在检索时消费而非进入重排提示）、单次反思更新（无任何梯度步实现上下文内画像演化）。在四个公开基准（Yelp、Books、MovieTV、Goodreads）上，HARP 把 Hit@1 较最强已发布基线提升 6.9%~24.0%，在每个数据集上经 Holm 校正的 p < 1e-9。消融与逐案例归因显示三个算子的贡献互补，其中检索侧门是主要来源；我们也报告了反思步骤引入的权衡。代码已开源。

原文摘要

In LLM-based recommendation, memory acts as a user-level agent in a memory-augmented backbone. Yet users and items remain asymmetric: only the user is given a structured profile; it enters a bandwidth-limited reranker prompt; and it is frozen after warmup. We recast LLM-based recommendation as an agentic-communication process in four stages: BUILD a self-description, TRANSMIT it along the collaborative graph, UNDERSTAND the other side, and SYNC it from the next interaction. This formulation reveals three unresolved communication challenges: the item side never produces a comparable self-description (one-sided BUILD); profile-level signals enter only the bandwidth-limited reranker prompt rather than retrieval, where they would not compete for context tokens (prompt-trapped UNDERSTAND); and the user's description is frozen after warmup with no principled update rule (absent SYNC). We instantiate the missing stages as HARP (Homophily-aware Agentic Recommendation via dual Profiles), a training-free, three-operator addition to a frozen LLM backbone: a rule-extracted symmetric item profile (zero extra LLM calls), a channel-decomposable profile-level homophily H(phi_u, phi_i) consumed at retrieval time rather than inside the reranker prompt, and a single-pass reflective update realising in-context profile evolution without any gradient step. On four public benchmarks (Yelp, Books, MovieTV, Goodreads), HARP lifts Hit@1 over the strongest published baseline by 6.9%-24.0%, with family-wise Holm-corrected p < 1e-9 on every dataset. Ablations and a per-case attribution study show that the operators contribute complementary gains, with the retrieval-side gate as the main source; we also report the trade-off introduced by the reflective pass. We formulate agentic communication and provide a training-free instantiation. Code is available at https://github.com/Jove-Zhen/HARP.

---

*自动采集于 2026-11-05*

#论文 #arXiv #AI #小凯