@inproceedings{zheng-etal-2026-onevision,
title = "{O}ne{V}ision: An End-to-End Generative Framework for Multi-view {E}-commerce Vision Search",
author = "Zheng, Zexin and
Dai, Huangyu and
Mao, Lingtao and
Sun, Xinyu and
Liang, Zihan and
Chen, Ben and
Lei, Chenyi",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EMNLP} 2026",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-emnlp.755/",
pages = "15895--15914",
ISBN = "979-8-89176-462-0",
abstract = "Traditional vision search follows the multi-stage cascading architecture (MCA) paradigm to balance efficiency and conversion, yet optimization objectives collide across stages, making it difficult to achieve Pareto optimality. Meanwhile, pronounced multi-view discrepancies of the same item undermine retrieval accuracy, as embedding-similarity methods fail to bridge large appearance variations. In this paper, we propose OneVision, an end-to-end generative framework that addresses both challenges. OneVision introduces VRQ, a vision-aligned residual quantization method that assigns consistent shallow-level codes to visually similar images of the same item while preserving fine-grained differences at deeper levels. It further employs a multi-stage training pipeline to retrieve all relevant items across views for a query image and incorporates user-specific signals for personalized preference modeling. In offline evaluations, OneVision performs on par with online MCA. Dynamic pruning further reduces its inference latency by 18{\%} over the unpruned variant. In A/B tests, it achieves significant improvements: +3.61{\%} item CTR, +2.43{\%} CVR, and +5.89{\%} order volume. The code will be publicly available."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zheng-etal-2026-onevision">
<titleInfo>
<title>OneVision: An End-to-End Generative Framework for Multi-view E-commerce Vision Search</title>
</titleInfo>
<name type="personal">
<namePart type="given">Zexin</namePart>
<namePart type="family">Zheng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Huangyu</namePart>
<namePart type="family">Dai</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lingtao</namePart>
<namePart type="family">Mao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xinyu</namePart>
<namePart type="family">Sun</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zihan</namePart>
<namePart type="family">Liang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ben</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chenyi</namePart>
<namePart type="family">Lei</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-462-0</identifier>
</relatedItem>
<abstract>Traditional vision search follows the multi-stage cascading architecture (MCA) paradigm to balance efficiency and conversion, yet optimization objectives collide across stages, making it difficult to achieve Pareto optimality. Meanwhile, pronounced multi-view discrepancies of the same item undermine retrieval accuracy, as embedding-similarity methods fail to bridge large appearance variations. In this paper, we propose OneVision, an end-to-end generative framework that addresses both challenges. OneVision introduces VRQ, a vision-aligned residual quantization method that assigns consistent shallow-level codes to visually similar images of the same item while preserving fine-grained differences at deeper levels. It further employs a multi-stage training pipeline to retrieve all relevant items across views for a query image and incorporates user-specific signals for personalized preference modeling. In offline evaluations, OneVision performs on par with online MCA. Dynamic pruning further reduces its inference latency by 18% over the unpruned variant. In A/B tests, it achieves significant improvements: +3.61% item CTR, +2.43% CVR, and +5.89% order volume. The code will be publicly available.</abstract>
<identifier type="citekey">zheng-etal-2026-onevision</identifier>
<location>
<url>https://aclanthology.org/2026.findings-emnlp.755/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>15895</start>
<end>15914</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T OneVision: An End-to-End Generative Framework for Multi-view E-commerce Vision Search
%A Zheng, Zexin
%A Dai, Huangyu
%A Mao, Lingtao
%A Sun, Xinyu
%A Liang, Zihan
%A Chen, Ben
%A Lei, Chenyi
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F zheng-etal-2026-onevision
%X Traditional vision search follows the multi-stage cascading architecture (MCA) paradigm to balance efficiency and conversion, yet optimization objectives collide across stages, making it difficult to achieve Pareto optimality. Meanwhile, pronounced multi-view discrepancies of the same item undermine retrieval accuracy, as embedding-similarity methods fail to bridge large appearance variations. In this paper, we propose OneVision, an end-to-end generative framework that addresses both challenges. OneVision introduces VRQ, a vision-aligned residual quantization method that assigns consistent shallow-level codes to visually similar images of the same item while preserving fine-grained differences at deeper levels. It further employs a multi-stage training pipeline to retrieve all relevant items across views for a query image and incorporates user-specific signals for personalized preference modeling. In offline evaluations, OneVision performs on par with online MCA. Dynamic pruning further reduces its inference latency by 18% over the unpruned variant. In A/B tests, it achieves significant improvements: +3.61% item CTR, +2.43% CVR, and +5.89% order volume. The code will be publicly available.
%U https://aclanthology.org/2026.findings-emnlp.755/
%P 15895-15914
Markdown (Informal)
[OneVision: An End-to-End Generative Framework for Multi-view E-commerce Vision Search](https://aclanthology.org/2026.findings-emnlp.755/) (Zheng et al., Findings 2026)
ACL