问HN：2024-2025年有哪些推荐系统的论文？
我一直在关注经典文献（如NCF、Wide & Deep、LightGCN），但在过去18到24个月里，领域似乎发生了剧烈变化，转向基于大语言模型（LLM）的推理和大规模图检索。
我在寻找2026年的“最先进”技术。具体来说：
- LLM4Rec：不仅仅是将LLM用于特征工程——谁在生成推荐方面做得很好？
- 检索与排序：在“两塔”范式或向量数据库集成方面有新的突破吗？
- 现实世界规模：有哪些论文讨论了这些新型重模型的延迟/成本权衡？
- 最近你读过的、对你思考发现方式影响最大的论文是什么？
查看原文
I’ve been keeping up with the classics (NCF, Wide & Deep, LightGCN), but the field seems to have shifted dramatically in the last 18–24 months toward LLM-based reasoning and graph-based retrieval at scale.<p>I’m looking for the "state of the art" in 2026. Specifically:<p>LLM4Rec: Beyond just using LLMs for feature engineering—who is doing generative recommendation well?<p>Retrieval vs. Ranking: Any new breakthroughs in the "Two-Tower" paradigm or vector database integration?<p>Real-world Scale: Papers that address the latency/cost trade-offs of these newer, heavier models.<p>What has been the most influential paper you’ve read recently that changed how you think about discovery?