新闻
HARP: Training-Free Dual-Profile Agentic Communication for LLM-Based Recommendation
arXiv cs.AI · 发布于 · 阅读约3分钟
30秒读懂
- 发生了什么
- HARP adds three operators to frozen LLM recommendation systems: symmetric item profiles, retrieval-level homophily matching, and reflective profile updates without retraining.
- 为何重要
- Engineers building LLM-based recommenders who want better ranking accuracy without modifying the base model or retraining.
- 注意
- The method requires access to item metadata for profile extraction and adds computational cost during retrieval; gains vary across datasets.
- agent
- agentic
- llm
- reranker
- prompt
这条新闻背后的模式
每个模式都讲清楚技术如何运作、何时值得投入，以及在哪里会失效。
The Agent Architect
每周一个模式、一个权衡、一个生产事故案例。为构建智能体系统的人准备的每周简报。
每周一封邮件,一键退订。您的地址仅用于发送简报。