Saved in:
Bibliographic Details
Main Authors: Yuan, Xinhang, Huang, Zexi, Cao, Anjia, Lu, Xudong, Wang, Zikai, Zhou, Penghao, Liu, Chang, Guo, Wentao, Wang, Qinglei
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2605.21832
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913165267173376
author Yuan, Xinhang
Huang, Zexi
Cao, Anjia
Lu, Xudong
Wang, Zikai
Zhou, Penghao
Liu, Chang
Guo, Wentao
Wang, Qinglei
author_facet Yuan, Xinhang
Huang, Zexi
Cao, Anjia
Lu, Xudong
Wang, Zikai
Zhou, Penghao
Liu, Chang
Guo, Wentao
Wang, Qinglei
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2605_21832
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle FLUID: From Ephemeral IDs to Multimodal Semantic Codes for Industrial-Scale Livestreaming Recommendation
Yuan, Xinhang
Huang, Zexi
Cao, Anjia
Lu, Xudong
Wang, Zikai
Zhou, Penghao
Liu, Chang
Guo, Wentao
Wang, Qinglei
Artificial Intelligence
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.
title FLUID: From Ephemeral IDs to Multimodal Semantic Codes for Industrial-Scale Livestreaming Recommendation
topic Artificial Intelligence
url https://arxiv.org/abs/2605.21832