Reproduce “IntentGR: Enhancing Intent for End-to-End Personalized Generative Retrieval in Search” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Gefan Ye, Yanjing Wu, Yinfu Feng, Zhiqiang Wang, Liang Li, Jun Xiao, Rong Xiao, Xiaoyi Zeng

IntentGR combines query intent and behavior in end-to-end generative retrieval, generating semantic item identifiers rather than ranking candidates. It adds clicks from others who used the same query, aligns query-item learning, enriches long-tail query training, and guides decoding with query information. Against query-aware ETEGRec-Q, Recall@10 rises from 69.88% to 77.99% on AOL4PS and from 22.40% to 34.36% on AE, in offline benchmarks.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.