Learning User Interests via Reasoning and Distillation for Cross-Domain News Recommendation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhu, Mengdan, Zhao, Yufan, Di, Tao, Yan, Yulan, Zhao, Liang
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915800402624512
author Zhu, Mengdan
Zhao, Yufan
Di, Tao
Yan, Yulan
Zhao, Liang
author_facet Zhu, Mengdan
Zhao, Yufan
Di, Tao
Yan, Yulan
Zhao, Liang
contents News recommendation plays a critical role in online news platforms by helping users discover relevant content. Cross-domain news recommendation further requires inferring user's underlying information needs from heterogeneous signals that often extend beyond direct news consumption. A key challenge lies in moving beyond surface-level behaviors to capture deeper, reusable user interests while maintaining scalability in large-scale production systems. In this paper, we present a reinforcement learning framework that trains large language models to generate high-quality lists of interest-driven news search queries from cross-domain user signals. We formulate query-list generation as a policy optimization problem and employ GRPO with multiple reward signals. We systematically study two compute dimensions: inference-time sampling and model capacity, and empirically observe consistent improvements with increased compute that exhibit scaling-like behavior. Finally, we perform on-policy distillation to transfer the learned policy from a large, compute-intensive teacher to a compact student model suitable for scalable deployment. Extensive offline experiments, ablation studies and large-scale online A/B tests in a production news recommendation system demonstrate consistent gains in both interest modeling quality and downstream recommendation performance.
format Preprint
id arxiv_https___arxiv_org_abs_2602_15005
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Learning User Interests via Reasoning and Distillation for Cross-Domain News Recommendation
Zhu, Mengdan
Zhao, Yufan
Di, Tao
Yan, Yulan
Zhao, Liang
Computation and Language
Information Retrieval
News recommendation plays a critical role in online news platforms by helping users discover relevant content. Cross-domain news recommendation further requires inferring user's underlying information needs from heterogeneous signals that often extend beyond direct news consumption. A key challenge lies in moving beyond surface-level behaviors to capture deeper, reusable user interests while maintaining scalability in large-scale production systems. In this paper, we present a reinforcement learning framework that trains large language models to generate high-quality lists of interest-driven news search queries from cross-domain user signals. We formulate query-list generation as a policy optimization problem and employ GRPO with multiple reward signals. We systematically study two compute dimensions: inference-time sampling and model capacity, and empirically observe consistent improvements with increased compute that exhibit scaling-like behavior. Finally, we perform on-policy distillation to transfer the learned policy from a large, compute-intensive teacher to a compact student model suitable for scalable deployment. Extensive offline experiments, ablation studies and large-scale online A/B tests in a production news recommendation system demonstrate consistent gains in both interest modeling quality and downstream recommendation performance.
title Learning User Interests via Reasoning and Distillation for Cross-Domain News Recommendation
topic Computation and Language
Information Retrieval
url https://arxiv.org/abs/2602.15005