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

Modern techniques in Content-based Recommendation (CBR) leverage item contentinformation to provide personalized services to users, but suffer fromresource-intensive training on large datasets. To address this issue, weexplore the dataset condensation for textual CBR in this paper. The goal ofdataset condensation is to synthesize a small yet informative dataset, uponwhich models can achieve performance comparable to those trained on largedatasets. While existing condensation approaches are tailored to classificationtasks for continuous data like images or embeddings, direct application of themto CBR has limitations. To bridge this gap, we investigate efficient datasetcondensation for content-based recommendation. Inspired by the remarkableabilities of large language models (LLMs) in text comprehension and generation,we leverage LLMs to empower the generation of textual content duringcondensation. To handle the interaction data involving both users and items, wedevise a dual-level condensation method: content-level and user-level. Atcontent-level, we utilize LLMs to condense all contents of an item into a newinformative title. At user-level, we design a clustering-based synthesismodule, where we first utilize LLMs to extract user interests. Then, the userinterests and user embeddings are incorporated to condense users and generateinteractions for condensed users. Notably, the condensation paradigm of thismethod is forward and free from iterative optimization on the synthesizeddataset. Extensive empirical findings from our study, conducted on threeauthentic datasets, substantiate the efficacy of the proposed method.Particularly, we are able to approximate up to 97% of the original performancewhile reducing the dataset size by 95% (i.e., on dataset MIND).

TF-DCon: Leveraging Large Language Models (LLMs) to Empower Training-Free Dataset Condensation for Content-Based Recommendation