Q-BERT4Rec: Quantized Semantic-ID Representation Learning for Multimodal Recommendation

Haofeng Huang 1, Ling Gai 2,* · 2025

Read Q-BERT4Rec: Quantized Semantic-ID Representation Learning for Multimodal Recommendation by Haofeng Huang 1, Ling Gai 2,* (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2512-02474.txt.

Description

Haofeng Huang and Ling Gai study how online shops guess what a user will pick next. Sequential recommendation drives personal picks on shopping, ad, and streaming sites. Strong recent systems, such as BERT4Rec, lean on plain item numbers that carry no meaning, and they ignore text and images.

Their 2025 study, called Q-BERT4Rec, gives each item a compact code built from its meaning instead. Text and image details feed into that code, so similar items sit close together. Richer item codes could help shops suggest goods that fit taste, even for new or rare items. Good codes save shops from blind guesses when a user has odd taste in goods.

Catalog record

CREATORHaofeng Huang 1, Ling Gai 2,*
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE48 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2512-02474

Browse the full library · About the Tower of Babel