@inproceedings{liu-etal-2026-casg,
title = "{CASG}: Drive Text to {ID}s for Sequential Recommendation via Collaboration-Aware Spectral Guidance",
author = "Liu, Yichen and
Zhang, Lianyue and
Xu, Cong and
Zhu, Zhangchi and
Chen, Yanyu and
He, Yunhang and
Wang, Jun and
Zhang, Wei",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-main.2001/",
pages = "45588--45602",
ISBN = "979-8-89176-459-0",
abstract = "Text embeddings incorporate additional rich semantic information; however, their performance in sequential recommendation still falls short of that of ID embeddings. Decomposing item embeddings along the spectrum of the collaborative item{--}item graph reveals a distinct discrepancy in their spectral energy distribution. Task-learned ID embeddings concentrate their energy heavily in the low- and mid-frequency directions of the collaborative graph topology, while textual representations spread more uniformly across the entire spectrum. Motivated by this insight, we propose CASG, a learnable polynomial graph filter that shifts text embeddings toward ID embeddings via collaboration-aware spectral guidance. Specifically, it leverages collaborative relationships to actively guide and filter the spectral energy distribution of frozen text embeddings, selectively preserving the frequency components aligned with the collaborative topology in a single offline pass. Across three datasets and two text encoders, CASG consistently surpasses learned ID embeddings, alignment-based, and fusion-based baselines at no online cost, suggesting that what semantic representations need is neither more alignment nor more fusion but collaboration-aware spectral guidance. Our source code is available at https://github.com/blueColdSea/CASG."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="liu-etal-2026-casg">
<titleInfo>
<title>CASG: Drive Text to IDs for Sequential Recommendation via Collaboration-Aware Spectral Guidance</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yichen</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lianyue</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Cong</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhangchi</namePart>
<namePart type="family">Zhu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yanyu</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yunhang</namePart>
<namePart type="family">He</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jun</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wei</namePart>
<namePart type="family">Zhang</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>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</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">979-8-89176-459-0</identifier>
</relatedItem>
<abstract>Text embeddings incorporate additional rich semantic information; however, their performance in sequential recommendation still falls short of that of ID embeddings. Decomposing item embeddings along the spectrum of the collaborative item–item graph reveals a distinct discrepancy in their spectral energy distribution. Task-learned ID embeddings concentrate their energy heavily in the low- and mid-frequency directions of the collaborative graph topology, while textual representations spread more uniformly across the entire spectrum. Motivated by this insight, we propose CASG, a learnable polynomial graph filter that shifts text embeddings toward ID embeddings via collaboration-aware spectral guidance. Specifically, it leverages collaborative relationships to actively guide and filter the spectral energy distribution of frozen text embeddings, selectively preserving the frequency components aligned with the collaborative topology in a single offline pass. Across three datasets and two text encoders, CASG consistently surpasses learned ID embeddings, alignment-based, and fusion-based baselines at no online cost, suggesting that what semantic representations need is neither more alignment nor more fusion but collaboration-aware spectral guidance. Our source code is available at https://github.com/blueColdSea/CASG.</abstract>
<identifier type="citekey">liu-etal-2026-casg</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.2001/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>45588</start>
<end>45602</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T CASG: Drive Text to IDs for Sequential Recommendation via Collaboration-Aware Spectral Guidance
%A Liu, Yichen
%A Zhang, Lianyue
%A Xu, Cong
%A Zhu, Zhangchi
%A Chen, Yanyu
%A He, Yunhang
%A Wang, Jun
%A Zhang, Wei
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F liu-etal-2026-casg
%X Text embeddings incorporate additional rich semantic information; however, their performance in sequential recommendation still falls short of that of ID embeddings. Decomposing item embeddings along the spectrum of the collaborative item–item graph reveals a distinct discrepancy in their spectral energy distribution. Task-learned ID embeddings concentrate their energy heavily in the low- and mid-frequency directions of the collaborative graph topology, while textual representations spread more uniformly across the entire spectrum. Motivated by this insight, we propose CASG, a learnable polynomial graph filter that shifts text embeddings toward ID embeddings via collaboration-aware spectral guidance. Specifically, it leverages collaborative relationships to actively guide and filter the spectral energy distribution of frozen text embeddings, selectively preserving the frequency components aligned with the collaborative topology in a single offline pass. Across three datasets and two text encoders, CASG consistently surpasses learned ID embeddings, alignment-based, and fusion-based baselines at no online cost, suggesting that what semantic representations need is neither more alignment nor more fusion but collaboration-aware spectral guidance. Our source code is available at https://github.com/blueColdSea/CASG.
%U https://aclanthology.org/2026.emnlp-main.2001/
%P 45588-45602
Markdown (Informal)
[CASG: Drive Text to IDs for Sequential Recommendation via Collaboration-Aware Spectral Guidance](https://aclanthology.org/2026.emnlp-main.2001/) (Liu et al., EMNLP 2026)
ACL
- Yichen Liu, Lianyue Zhang, Cong Xu, Zhangchi Zhu, Yanyu Chen, Yunhang He, Jun Wang, and Wei Zhang. 2026. CASG: Drive Text to IDs for Sequential Recommendation via Collaboration-Aware Spectral Guidance. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 45588–45602, Budapest, Hungary. Association for Computational Linguistics.