English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Recommendation models that utilize unique identities (IDs) to representdistinct users and items have been state-of-the-art (SOTA) and dominated therecommender systems (RS) literature for over a decade. Meanwhile, thepre-trained modality encoders, such as BERT and ViT, have become increasinglypowerful in modeling the raw modality features of an item, such as text andimages. Given this, a natural question arises: can a purely modality-basedrecommendation model (MoRec) outperforms or matches a pure ID-based model(IDRec) by replacing the itemID embedding with a SOTA modality encoder? Infact, this question was answered ten years ago when IDRec beats MoRec by astrong margin in both recommendation accuracy and efficiency. We aim to revisitthis `old' question and systematically study MoRec from several aspects.Specifically, we study several sub-questions: (i) which recommendationparadigm, MoRec or IDRec, performs better in practical scenarios, especially inthe general setting and warm item scenarios where IDRec has a strong advantage?does this hold for items with different modality features? (ii) can the latesttechnical advances from other communities (i.e., natural language processingand computer vision) translate into accuracy improvement for MoRec? (iii) howto effectively utilize item modality representation, can we use it directly ordo we have to adjust it with new data? (iv) are there some key challenges forMoRec to be solved in practical applications? To answer them, we conductrigorous experiments for item recommendations with two popular modalities,i.e., text and vision. We provide the first empirical evidence that MoRec isalready comparable to its IDRec counterpart with an expensive end-to-endtraining method, even for warm item recommendation. Our results potentiallyimply that the dominance of IDRec in the RS field may be greatly challenged inthe future.

Where to Go Next for Recommender Systems? ID- vs. Modality-based Recommender Models Revisited