Tsinghua University Wins SIGIR 2024 "Test of Time Paper Award" and "Best Paper Award" for the First Time as a Mainland Chinese Research Institution

The 47th annual ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2024) recently announced its award winners. Among the recipients, teams from Tsinghua University's Department of Computer Science received two major awards:

  1. The Test of Time Award was given to a 2014 paper titled "Explicit factor models for explainable recommendation based on phrase-level sentiment analysis". This paper was authored by Zhang Yongfeng, Lai Guokun, and others under the guidance of Professors Zhang Min, Liu Yiqun, and Ma Shaoping. The research defined the problem of "explainable recommendation" for the first time and designed corresponding sentiment analysis algorithms to address this technical challenge. Since its publication, it has played a leading role in the design and implementation of internet recommendation systems.

  2. The Best Paper Award went to "Scaling Laws For Dense Retrieval" by Fang Yan, Zhan Jingtao, and supervised by Assistant Professor Ai Qingyao and Professor Liu Yiqun. This is the first time a research institution from mainland China has led a paper winning this award. The study investigated the applicability of scaling laws in dense information retrieval, which has important implications for the design of search engines, recommendation systems, and other information retrieval systems.

Additionally, Assistant Professor Ai Qingyao from Tsinghua's CS department received the Early Career Award, marking the first time a researcher from mainland China has won this award.

Other award recipients include:

  • Early Career Award: Bhaskar Mitra (Microsoft Research), Harrie Oosterhuis (Radboud University), and Xiang Wang (University of Science and Technology of China)
  • Community Awards and DEI Awards were also presented to three recipients

The conference received 1148 submissions, with 791 valid submissions and 159 full papers accepted, resulting in an acceptance rate of 20.1%.

A paper titled "Generative Retrieval as Multi-Vector Dense Retrieval" by researchers from Shandong University, Leiden University, and the University of Amsterdam received a Best Paper Honorable Mention.

[Links to papers: