@inproceedings{shi-etal-2026-llada,
title = "{LL}a{DA}-Rec: Discrete Diffusion for Parallel Semantic {ID} Generation in Generative Recommendation",
author = "Shi, Teng and
Shen, Chenglei and
Yu, Weijie and
Nie, Shen and
Li, Chongxuan and
Zhang, Xiao and
He, Ming and
Fan, Jianping and
Xu, Jun",
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.616/",
pages = "14152--14165",
ISBN = "979-8-89176-459-0",
abstract = "Generative recommendation represents items as semantic identifiers (SIDs), i.e., sequences of discrete tokens, and typically predict the next item autoregressively. Despite their strong performance, such methods suffer from two limitations: (1) unidirectional constraints, where causal attention enforces left-to-right dependencies and limits global semantic modeling; and (2) error accumulation, where early mistakes propagate to later tokens and hurt recommendation accuracy. To address these issues, we propose LLaDA-Rec, a discrete diffusion framework that reformulates next-item recommendation as parallel SID generation. By combining bidirectional attention with an adaptive generation order, LLaDA-Rec captures richer inter-item and intra-item dependencies while reducing error accumulation. Our framework includes three key components: (1) a parallel tokenization mechanism for diffusion-compatible SIDs; (2) masking strategies at both the user-history and next-item levels; and (3) an adapted beam search algorithm for adaptive-order discrete diffusion decoding. Experiments on three public datasets show that LLaDA-Rec consistently outperforms both conventional ID-based recommenders and strong generative baselines, highlighting discrete diffusion as a promising paradigm for generative recommendation."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="shi-etal-2026-llada">
<titleInfo>
<title>LLaDA-Rec: Discrete Diffusion for Parallel Semantic ID Generation in Generative Recommendation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Teng</namePart>
<namePart type="family">Shi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chenglei</namePart>
<namePart type="family">Shen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Weijie</namePart>
<namePart type="family">Yu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shen</namePart>
<namePart type="family">Nie</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chongxuan</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiao</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ming</namePart>
<namePart type="family">He</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jianping</namePart>
<namePart type="family">Fan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jun</namePart>
<namePart type="family">Xu</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>Generative recommendation represents items as semantic identifiers (SIDs), i.e., sequences of discrete tokens, and typically predict the next item autoregressively. Despite their strong performance, such methods suffer from two limitations: (1) unidirectional constraints, where causal attention enforces left-to-right dependencies and limits global semantic modeling; and (2) error accumulation, where early mistakes propagate to later tokens and hurt recommendation accuracy. To address these issues, we propose LLaDA-Rec, a discrete diffusion framework that reformulates next-item recommendation as parallel SID generation. By combining bidirectional attention with an adaptive generation order, LLaDA-Rec captures richer inter-item and intra-item dependencies while reducing error accumulation. Our framework includes three key components: (1) a parallel tokenization mechanism for diffusion-compatible SIDs; (2) masking strategies at both the user-history and next-item levels; and (3) an adapted beam search algorithm for adaptive-order discrete diffusion decoding. Experiments on three public datasets show that LLaDA-Rec consistently outperforms both conventional ID-based recommenders and strong generative baselines, highlighting discrete diffusion as a promising paradigm for generative recommendation.</abstract>
<identifier type="citekey">shi-etal-2026-llada</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.616/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>14152</start>
<end>14165</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T LLaDA-Rec: Discrete Diffusion for Parallel Semantic ID Generation in Generative Recommendation
%A Shi, Teng
%A Shen, Chenglei
%A Yu, Weijie
%A Nie, Shen
%A Li, Chongxuan
%A Zhang, Xiao
%A He, Ming
%A Fan, Jianping
%A Xu, Jun
%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 shi-etal-2026-llada
%X Generative recommendation represents items as semantic identifiers (SIDs), i.e., sequences of discrete tokens, and typically predict the next item autoregressively. Despite their strong performance, such methods suffer from two limitations: (1) unidirectional constraints, where causal attention enforces left-to-right dependencies and limits global semantic modeling; and (2) error accumulation, where early mistakes propagate to later tokens and hurt recommendation accuracy. To address these issues, we propose LLaDA-Rec, a discrete diffusion framework that reformulates next-item recommendation as parallel SID generation. By combining bidirectional attention with an adaptive generation order, LLaDA-Rec captures richer inter-item and intra-item dependencies while reducing error accumulation. Our framework includes three key components: (1) a parallel tokenization mechanism for diffusion-compatible SIDs; (2) masking strategies at both the user-history and next-item levels; and (3) an adapted beam search algorithm for adaptive-order discrete diffusion decoding. Experiments on three public datasets show that LLaDA-Rec consistently outperforms both conventional ID-based recommenders and strong generative baselines, highlighting discrete diffusion as a promising paradigm for generative recommendation.
%U https://aclanthology.org/2026.emnlp-main.616/
%P 14152-14165
Markdown (Informal)
[LLaDA-Rec: Discrete Diffusion for Parallel Semantic ID Generation in Generative Recommendation](https://aclanthology.org/2026.emnlp-main.616/) (Shi et al., EMNLP 2026)
ACL
- Teng Shi, Chenglei Shen, Weijie Yu, Shen Nie, Chongxuan Li, Xiao Zhang, Ming He, Jianping Fan, and Jun Xu. 2026. LLaDA-Rec: Discrete Diffusion for Parallel Semantic ID Generation in Generative Recommendation. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 14152–14165, Budapest, Hungary. Association for Computational Linguistics.