Junguang Jiang, Yanwen Huang, Liu Bin, Xiaoyu Kong, Xinhang Li, Ziru Xu, Han Zhu, Jian Xu, Bo Zheng · ACM International Conference on Information and Knowledge Management (CIKM) 2026 · 2026
DOI: 10.1145/3799682.3840090
Counts differ because each database indexes a different set of publications. We treat OpenAlex as the canonical count; Google Scholar is not shown (no API, and crawling it violates its ToS).
In real-world recommender systems, different retrieval objectives are typically addressed using task-specific datasets with carefully designed model architectures. We demonstrate that Large Language Models (LLMs) can function as universal retrievers, capable of handling multiple objectives within a generative retrieval framework. To address the model expressiveness and inference efficiency issues of generative retrieval, we propose multi-query representation and transferable discriminative mapping layers to model complex user-item relationships, and incorporate probabilistic sampling to reduce computation costs. Finally, our Universal Retrieval Model (URM) can adaptively generate a set from tens of millions of candidates based on an arbitrary given objective while keeping the latency within tens of milliseconds. Applied to industrial-scale data, URM outperforms expert models elaborately designed for different retrieval objectives in offline experiments and achieves a 3.13% revenue increase on our online advertising platform. URM also demonstrates strong performance in impression and GMV for long-tail and cold-start items.
Last synced
This work has 20 recorded citations, but citing papers have not been linked locally yet.
No comments yet — start the discussion below.