Keigo Sakurai, Ph.D.

Assistant Professor, Faculty of Information Science and Technology / D-RED, Hokkaido University

KS.jpg

Laboratory of Media Creation Methodology

Hokkaido University

Sapporo, Japan

sakurai [at] lmd.ist.hokudai.ac.jp

I am an Assistant Professor at Hokkaido University, working with the Laboratory of Media Creation Methodology. I received my Ph.D. in Information Science from Hokkaido University in September 2025, where I was a JSPS Research Fellow (DC1).

My research focuses on data mining centered on recommender systems, applied across multiple domains including music, finance, and fashion. Currently, I am particularly interested in graph-based recommendation, sequential recommendation, generative retrieval and recommendation, explainability, and the use of large language models for recommendation in various domains.

My work has been published in premiere venues such as SIGIR, RecSys, WSDM, CIKM, ECIR, ISMIR, and ICAIF. I am a member of ACM, IEEE, the Japanese Society for Artificial Intelligence (JSAI), and the Information Processing Society of Japan (IPSJ).

櫻井 慶悟(さくらい けいご)の日本語プロフィールはこちら。

news

Nov 03, 2026 One paper on the reach of recommendation is accepted to WSDM 2027 (Findings).
Oct 21, 2026 Our RecSys 2026 work is featured in Nikkei (The Nikkei, Japan’s leading business daily).
Oct 03, 2026 One paper on decorrelated re-ranking for financial asset recommendation is accepted to ICAIF 2026.
Aug 11, 2026 Two papers on graph collaborative filtering and sequential recommendation are accepted to CIKM 2026.
Jul 11, 2026 One paper on elucidating the last item-reliance effects in sequential recommendation is accepted to RecSys 2026.

selected publications

  1. WSDM
    The Illusion of Reach in Recommendation
    Keigo Sakurai, Takahiro Ogawa, Jong-June Jeon, and 1 more author
    In Proceedings of the ACM International Conference on Web Search and Data Mining (WSDM) (Findings, acceptance rate: 12% for Findings, 15% for Main Conference), 2027
  2. CIKM
    The Edge Spectrum of Choice-Derived Item Graphs: Strong and Weak Edges Encode Different Relations in Collaborative Filtering
    Keigo Sakurai, Takahiro Ogawa, and Miki Haseyama
    In Proceedings of the ACM International Conference on Information and Knowledge Management (CIKM) (Full Paper, acceptance rate: 27%), 2026
  3. RecSys
    Residual Dominance as a Structural Account of Last-Item Reliance in Causal Self-Attention Recommenders
    Keito Kozaki, Keigo Sakurai, Ren Togo, and 2 more authors
    In Proceedings of the ACM Conference on Recommender Systems (RecSys) (Main Track, acceptance rate: 18%), 2026
  4. SIGIR
    Revisiting the Role of Learned Attention Weighting in SASRec
    Keito Kozaki, Keigo Sakurai, Ren Togo, and 2 more authors
    In Proceedings of the International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) (Short Paper Track, acceptance rate: 27%), 2026
  5. WSDM
    Risk-Aware Utility Re-Ranking for Financial Asset Recommendation
    Keigo Sakurai, Takahiro Ogawa, Miki Haseyama, and 2 more authors
    In Proceedings of the ACM International Conference on Web Search and Data Mining (WSDM) (Short Paper Track, acceptance rate: 35%), 2026
  6. ECIR
    LLM Is Knowledge Graph Reasoner: LLM’s Intuition-Aware Knowledge Graph Reasoning for Cold-Start Sequential Recommendation
    Keigo Sakurai, Ren Togo, Takahiro Ogawa, and 1 more author
    In Proceedings of the European Conference on Information Retrieval (ECIR) (Full Paper, acceptance rate: 23%), 2025
  7. ISMIR
    MMT-BERT: Chord-Aware Symbolic Music Generation Based on Multitrack Music Transformer and MusicBERT
    Jinlong Zhu, Keigo Sakurai, Ren Togo, and 2 more authors
    In Proceedings of the International Society for Music Information Retrieval Conference (ISMIR) (acceptance rate: 36%), 2024