@inproceedings{dinushka-etal-2026-clasp,
title = "{CLASP}: Collaborative Latent Alignment from Semantic Preference Descriptions for Natural Language based Recommendation",
author = "Dinushka, Dilan and
Ledent, Antoine and
Lauw, Hady W.",
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.1800/",
pages = "41013--41026",
ISBN = "979-8-89176-459-0",
abstract = "We propose CLASP, a lightweight model which performs recommendations for cold-start users for which only textual user description is provided and no previous interactions are observed. Our architecture learns lightweight latent alignment functions that project textual preference representations into a pretrained collaborative filtering latent space. The CF component can be pretrained on a large set of dense interactions from warm-start users, whilst the alignment maybe trained on a distinct set of users for which textual descriptions are present, allowing an efficient use of heterogeneous information. Extensive experiments on Steam, MovieLens-1M, and Netflix demonstrate that CLASP consistently outperforms retrieval-oriented and encoder-only semantic recommendation baselines while remaining substantially more computationally efficient than LLM-based recommendation methods. Remarkably, this happens despite the fact that some of the baselines have access to item text descriptions, whilst CLASP doesn{'}t. Lastly, we also provide a dataset which allows a granular study of the variability in user profile generation from small subsets of interactions."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="dinushka-etal-2026-clasp">
<titleInfo>
<title>CLASP: Collaborative Latent Alignment from Semantic Preference Descriptions for Natural Language based Recommendation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Dilan</namePart>
<namePart type="family">Dinushka</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Antoine</namePart>
<namePart type="family">Ledent</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hady</namePart>
<namePart type="given">W</namePart>
<namePart type="family">Lauw</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>We propose CLASP, a lightweight model which performs recommendations for cold-start users for which only textual user description is provided and no previous interactions are observed. Our architecture learns lightweight latent alignment functions that project textual preference representations into a pretrained collaborative filtering latent space. The CF component can be pretrained on a large set of dense interactions from warm-start users, whilst the alignment maybe trained on a distinct set of users for which textual descriptions are present, allowing an efficient use of heterogeneous information. Extensive experiments on Steam, MovieLens-1M, and Netflix demonstrate that CLASP consistently outperforms retrieval-oriented and encoder-only semantic recommendation baselines while remaining substantially more computationally efficient than LLM-based recommendation methods. Remarkably, this happens despite the fact that some of the baselines have access to item text descriptions, whilst CLASP doesn’t. Lastly, we also provide a dataset which allows a granular study of the variability in user profile generation from small subsets of interactions.</abstract>
<identifier type="citekey">dinushka-etal-2026-clasp</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.1800/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>41013</start>
<end>41026</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T CLASP: Collaborative Latent Alignment from Semantic Preference Descriptions for Natural Language based Recommendation
%A Dinushka, Dilan
%A Ledent, Antoine
%A Lauw, Hady W.
%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 dinushka-etal-2026-clasp
%X We propose CLASP, a lightweight model which performs recommendations for cold-start users for which only textual user description is provided and no previous interactions are observed. Our architecture learns lightweight latent alignment functions that project textual preference representations into a pretrained collaborative filtering latent space. The CF component can be pretrained on a large set of dense interactions from warm-start users, whilst the alignment maybe trained on a distinct set of users for which textual descriptions are present, allowing an efficient use of heterogeneous information. Extensive experiments on Steam, MovieLens-1M, and Netflix demonstrate that CLASP consistently outperforms retrieval-oriented and encoder-only semantic recommendation baselines while remaining substantially more computationally efficient than LLM-based recommendation methods. Remarkably, this happens despite the fact that some of the baselines have access to item text descriptions, whilst CLASP doesn’t. Lastly, we also provide a dataset which allows a granular study of the variability in user profile generation from small subsets of interactions.
%U https://aclanthology.org/2026.emnlp-main.1800/
%P 41013-41026
Markdown (Informal)
[CLASP: Collaborative Latent Alignment from Semantic Preference Descriptions for Natural Language based Recommendation](https://aclanthology.org/2026.emnlp-main.1800/) (Dinushka et al., EMNLP 2026)
ACL