Saved in:
Bibliographic Details
Main Authors: Zheng, Wenhao, Lu, Wang, Tang, Fangshuang, Lu, Yiyang, Yang, Jun, Xiong, Pengcheng, Yan, Yulan
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.00502
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915824684498944
author Zheng, Wenhao
Lu, Wang
Tang, Fangshuang
Lu, Yiyang
Yang, Jun
Xiong, Pengcheng
Yan, Yulan
author_facet Zheng, Wenhao
Lu, Wang
Tang, Fangshuang
Lu, Yiyang
Yang, Jun
Xiong, Pengcheng
Yan, Yulan
contents Early-stage users in a new scenario intensify cold-start challenges, yet prior works often address only parts of the problem through model architecture. Launching a new user experience to replace an established product involves sparse behavioral signals, low-engagement cohorts, and unstable model performance. We argue that effective recommendations require the synergistic integration of feature engineering, model architecture, and stable model updating. We propose Trinity, a framework embodying this principle. Trinity extracts valuable information from existing scenarios while ensuring predictive effectiveness and accuracy in the new scenario. In this paper, we showcase Trinity applied to a billion-user Microsoft product transition. Both offline and online experiments demonstrate that our framework achieves substantial improvements in addressing the combined challenge of new users in new scenarios.
format Preprint
id arxiv_https___arxiv_org_abs_2603_00502
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Trinity: A Scenario-Aware Recommendation Framework for Large-Scale Cold-Start Users
Zheng, Wenhao
Lu, Wang
Tang, Fangshuang
Lu, Yiyang
Yang, Jun
Xiong, Pengcheng
Yan, Yulan
Machine Learning
Early-stage users in a new scenario intensify cold-start challenges, yet prior works often address only parts of the problem through model architecture. Launching a new user experience to replace an established product involves sparse behavioral signals, low-engagement cohorts, and unstable model performance. We argue that effective recommendations require the synergistic integration of feature engineering, model architecture, and stable model updating. We propose Trinity, a framework embodying this principle. Trinity extracts valuable information from existing scenarios while ensuring predictive effectiveness and accuracy in the new scenario. In this paper, we showcase Trinity applied to a billion-user Microsoft product transition. Both offline and online experiments demonstrate that our framework achieves substantial improvements in addressing the combined challenge of new users in new scenarios.
title Trinity: A Scenario-Aware Recommendation Framework for Large-Scale Cold-Start Users
topic Machine Learning
url https://arxiv.org/abs/2603.00502