OnePiece: Bringing Context Engineering and Reasoning to Industrial Cascade Ranking System

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Dai, Sunhao, Tang, Jiakai, Wu, Jiahua, Wang, Kun, Zhu, Yuxuan, Chen, Bingjun, Hong, Bangyang, Zhao, Yu, Fu, Cong, Wu, Kangle, Ni, Yabo, Zeng, Anxiang, Wang, Wenjie, Chen, Xu, Xu, Jun, Ng, See-Kiong
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866911168899055616
author Dai, Sunhao
Tang, Jiakai
Wu, Jiahua
Wang, Kun
Zhu, Yuxuan
Chen, Bingjun
Hong, Bangyang
Zhao, Yu
Fu, Cong
Wu, Kangle
Ni, Yabo
Zeng, Anxiang
Wang, Wenjie
Chen, Xu
Xu, Jun
Ng, See-Kiong
author_facet Dai, Sunhao
Tang, Jiakai
Wu, Jiahua
Wang, Kun
Zhu, Yuxuan
Chen, Bingjun
Hong, Bangyang
Zhao, Yu
Fu, Cong
Wu, Kangle
Ni, Yabo
Zeng, Anxiang
Wang, Wenjie
Chen, Xu
Xu, Jun
Ng, See-Kiong
contents Despite the growing interest in replicating the scaled success of large language models (LLMs) in industrial search and recommender systems, most existing industrial efforts remain limited to transplanting Transformer architectures, which bring only incremental improvements over strong Deep Learning Recommendation Models (DLRMs). From a first principle perspective, the breakthroughs of LLMs stem not only from their architectures but also from two complementary mechanisms: context engineering, which enriches raw input queries with contextual cues to better elicit model capabilities, and multi-step reasoning, which iteratively refines model outputs through intermediate reasoning paths. However, these two mechanisms and their potential to unlock substantial improvements remain largely underexplored in industrial ranking systems. In this paper, we propose OnePiece, a unified framework that seamlessly integrates LLM-style context engineering and reasoning into both retrieval and ranking models of industrial cascaded pipelines. OnePiece is built on a pure Transformer backbone and further introduces three key innovations: (1) structured context engineering, which augments interaction history with preference and scenario signals and unifies them into a structured tokenized input sequence for both retrieval and ranking; (2) block-wise latent reasoning, which equips the model with multi-step refinement of representations and scales reasoning bandwidth via block size; (3) progressive multi-task training, which leverages user feedback chains to effectively supervise reasoning steps during training. OnePiece has been deployed in the main personalized search scenario of Shopee and achieves consistent online gains across different key business metrics, including over $+2\%$ GMV/UU and a $+2.90\%$ increase in advertising revenue.
format Preprint
id arxiv_https___arxiv_org_abs_2509_18091
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle OnePiece: Bringing Context Engineering and Reasoning to Industrial Cascade Ranking System
Dai, Sunhao
Tang, Jiakai
Wu, Jiahua
Wang, Kun
Zhu, Yuxuan
Chen, Bingjun
Hong, Bangyang
Zhao, Yu
Fu, Cong
Wu, Kangle
Ni, Yabo
Zeng, Anxiang
Wang, Wenjie
Chen, Xu
Xu, Jun
Ng, See-Kiong
Information Retrieval
Artificial Intelligence
Computation and Language
Despite the growing interest in replicating the scaled success of large language models (LLMs) in industrial search and recommender systems, most existing industrial efforts remain limited to transplanting Transformer architectures, which bring only incremental improvements over strong Deep Learning Recommendation Models (DLRMs). From a first principle perspective, the breakthroughs of LLMs stem not only from their architectures but also from two complementary mechanisms: context engineering, which enriches raw input queries with contextual cues to better elicit model capabilities, and multi-step reasoning, which iteratively refines model outputs through intermediate reasoning paths. However, these two mechanisms and their potential to unlock substantial improvements remain largely underexplored in industrial ranking systems. In this paper, we propose OnePiece, a unified framework that seamlessly integrates LLM-style context engineering and reasoning into both retrieval and ranking models of industrial cascaded pipelines. OnePiece is built on a pure Transformer backbone and further introduces three key innovations: (1) structured context engineering, which augments interaction history with preference and scenario signals and unifies them into a structured tokenized input sequence for both retrieval and ranking; (2) block-wise latent reasoning, which equips the model with multi-step refinement of representations and scales reasoning bandwidth via block size; (3) progressive multi-task training, which leverages user feedback chains to effectively supervise reasoning steps during training. OnePiece has been deployed in the main personalized search scenario of Shopee and achieves consistent online gains across different key business metrics, including over $+2\%$ GMV/UU and a $+2.90\%$ increase in advertising revenue.
title OnePiece: Bringing Context Engineering and Reasoning to Industrial Cascade Ranking System
topic Information Retrieval
Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2509.18091