English(EN) ChronicleRec: Pre-training Temporally Anchored Tokens for Lifelong User Modeling

ChronicleRec 框架压缩用户行为以改进推荐系统

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-11 02:47

研究人员开发了 ChronicleRec，一个用于预训练和迁移用户行为模型的新颖框架。该方法将超长历史行为序列压缩成按时间顺序排列的“Chronicle Tokens”，保留近期行为并简化远期历史。该框架使用因果编码器和多视界设计来学习互补的长期兴趣，并从旧历史中重建近期行为。在 KuaiRand 和 Tencent AdLive 数据集上的实验表明，ChronicleRec 的性能优于现有基线，并且为期七天的在线 A/B 测试证实了显著的生产收益。 AI

影响通过有效建模长期用户行为而不产生高昂的计算成本，提高了推荐系统的效率。

排序理由这是一篇详细介绍推荐系统中用户建模新方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.IR (Information Retrieval) 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 0 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 这是一篇详细介绍推荐系统中用户建模新方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, infra; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: 19 days old; Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.IR (Information Retrieval) TIER_1 English(EN) · Jie Jiang · 2026-09-11 02:47

ChronicleRec: 为终身用户建模预训练时间锚定Token

Modeling ultra-long user behavior sequences is crucial for industrial recommendation and online advertising, yet directly feeding thousands of historical actions into ranking models is computationally prohibitive, while truncation discards long-range signals. Existing lifelong-in…

本文如何被排名

报道来源 [1]

ChronicleRec: 为终身用户建模预训练时间锚定Token

相关实体

相关话题