SCoTER: Structured Chain-of-Thought Transfer for Enhanced Recommendation
Jie Jiang, Yang Wu, Qian Li, Yuling Xiong, Hongbo Tang, Xun Liu, Haoze Wang, Jun Zhang, Huan Yu, Hailong Shi

TL;DR
SCoTER introduces a unified framework that automates reasoning pattern discovery and preserves stepwise logic in LLM-based recommender systems, leading to improved performance and reduced inference costs.
Contribution
It proposes a novel joint optimization approach combining automated pattern discovery with structure-preserving transfer for recommender systems.
Findings
Consistent improvements on four benchmark datasets.
Achieved a 2.14% GMV lift in production deployment.
Eliminated online LLM inference costs.
Abstract
Harnessing the reasoning power of Large Language Models (LLMs) for recommender systems is hindered by two fundamental challenges. First, current approaches lack a mechanism for automated, data-driven discovery of effective reasoning patterns, relying instead on brittle manual templates or unstable zero-shot prompting. Second, they employ structure-collapsing integration: direct prompting incurs prohibitive online inference costs, while feature extraction collapses reasoning chains into single vectors, discarding stepwise logic. To address these challenges, we propose SCoTER (Structured Chain-of-Thought Transfer for Enhanced Recommendation), a unified framework that treats pattern discovery and structure-aware transfer as a jointly optimized problem. Specifically, SCoTER operationalizes this through two synergistic components: a Generate-Validate-Mine (GVM) pipeline for automated pattern…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsSentiment Analysis and Opinion Mining · Recommender Systems and Techniques · Topic Modeling
