用语义编码替代直播唯一ID，解决冷启动推荐难题。
FLUID: From Ephemeral IDs to Multimodal Semantic Codes for Industrial-Scale Livestreaming Recommendation
- 用多模态编码生成层级语义码，替代传统ID表征
- 上线后冷启动直播间观看量提升2.05%，总观看时长增0.55%
- 适合大规模直播推荐场景，尤其对短时内容有效
现代推荐系统依赖基于ID的协同过滤：每个商品由唯一ID嵌入表示，并累积用户交互信号。然而直播推荐面临独特挑战：一档直播仅持续数十分钟，导致其商品ID长期处于冷启动状态，以ID为中心的排序模型难以泛化。本文提出FLUID，首个在工业级直播推荐器中彻底摒弃候选侧商品ID的框架。FLUID引入跨域多模态编码器，联合训练于短视频与直播数据，生成离散分层语义码（称作LUCID），用于内容驱动的商品表征。为适配排序器，FLUID采用分阶段预热策略：先将冷启动的片段级LUCID作为独立令牌与ID嵌入并列使用，再在在线增量训练前以已热化的房间级LUCID替换原ID嵌入。该方案部署于覆盖全球超十亿用户的跨平台直播推荐系统，带来显著线上收益：+0.55%质量观看时长、+2.05%冷启动直播间浏览量、+0.05%活跃时长。
原文摘要 · Abstract (English)
Modern recommender systems rely heavily on ID-based collaborative filtering: each item is represented by a unique ID embedding that accumulates collaborative signals from user interactions. Livestreaming recommendation, however, faces a unique challenge in this paradigm: a live room typically broadcasts for only tens of minutes, so its item ID remains poorly learned in a persistent cold-start state and ID-centric ranking models fail to generalize. We present FLUID, the first framework to fully retire the candidate-side item ID from a production-scale livestreaming ranker. FLUID introduces a cross-domain multimodal encoder, jointly trained on short videos and livestreams, to produce discrete hierarchical semantic codes, called LUCID, for content-based item characterization. To adapt the ranker to LUCID, FLUID further employs a staged warmup scheme: it first incorporates cold, slice-level LUCID as an independent token alongside the ID embedding, and then replaces the ID embedding with warm, room-level LUCID before online incremental training. Deployed on our industrial livestreaming recommenders with a cross-platform combined user base of over one billion globally, FLUID delivers significant online gains of +0.55% Quality Watch Duration, +2.05% Cold-Start Room Views, and +0.05% Active Hours.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。