arXiv cs.AI· Renzhi Wu, Zikun Cui, Junjie Yang, Li Sheng, Yongjun Xie, Jerry Fu, Nipun Mathur, Sri Reddy, Hong Li·· 14 小时前AI 评分22
PRISM：用户历史中的个性化物品表示
PRISM: Personalized Item Representations in User History
AI 导读
PRISM 提出在用户历史中构建个性化物品表示，让同一物品的嵌入向量体现其与具体用户的关系，而非所有用户共享同一初始表示。该方法复用上游排序计算生成个性化物品嵌入，用自编码器将用户塔、用户—物品塔及同请求候选混合的各层状态压缩为每个打分对的一个嵌入。
正文 · 原文
Abstract:Sequential recommenders typically construct history inputs from item identifiers or content features using an embedding map shared across users. The same item therefore yields the same initial embedding representation, even when users have different relationships to the item. We propose personalized item representations in user history, where each item's embedding expresses its relationship to the user. Personalization becomes a property of the history inputs themselves, supplying richer evidence for later sequence modeling.
We instantiate this idea with PRISM, which reuses upstream ranking computation to generate personalized item embeddings. An autoencoder jointly compresses all layer states from a user tower, a user--item tower, and same-request candidate mixing into one embedding per scored pair for downstream replay. Offline evaluation on Surface 1, a major recommendation surface at billion scale, reports an average downstream normalized entropy (NE) change of $-0.15\%$, compared with $-0.07\%$ for VISTA at similar history lengths. We further identify an empirical sequence-length scaling law: over history lengths of 26--240 events, NE gain follows a fitted power law with exponent 0.468, increasing from 0.06\% to 0.17\%. Therefore, personalized item representations in user history provide a new scaling dimension for improving recommendation quality.
| Subjects: | Information Retrieval (cs.IR); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.39175 [cs.IR] |
| (or arXiv:2610.39175v1 [cs.IR] for this version) | |
| https://doi.org/10.48550/arXiv.2610.39175 arXiv-issued DOI via DataCite |
Submission history
From: Renzhi Wu [view email]
[v1]
Thu, 29 Oct 2026 02:05:57 UTC (118 KB)
来源：arXiv cs.AI · arxiv.org