Papers

You can also check my papers on Googlescholar.

Journal (Peer-reviewed)

1. On the Expectation of a Persistence Diagram by the Persistence Weighted Kernel, Genki Kusano. Japan Journal of Industrial and Applied Mathematics (JJIAM). Volume 36, pages 861–892. 2019. 

2. Kernel Method for Persistence Diagrams via Kernel Embedding and Weight Factor. Genki Kusano, Kenji Fukumizu, Yasuaki Hiraoka. Journal of Machine Learning Research (JMLR), 18(189):1--41, 2018. 

3. Relative interleavings and applications to sensor networks. Genki Kusano, Yasuaki Hiraoka. Japan Journal of Industrial and Applied Mathematics (JJIAM). 33(1):99-120, 2016. 

Proceedings (Peer-reviewed)

1. Revisiting Prompt Engineering: A Comprehensive Evaluation for LLM-based Personalized Recommendation. Genki Kusano, Kosuke Akimoto, Kunihiro Takeoka. 19th ACM Conference on Recommender Systems (RecSys). 832--841. 2025. [arXiv / GitHub]

2. Data Augmentation using Reverse Prompt for Cost-Efficient Cold-Start Recommendation. Genki Kusano. 18th ACM Conference on Recommender Systems (RecSys). 861--865. 2024. [Acceptance rate (short): 39/177=22.2%] 

3. GA-Tag: Data Enrichment with an Automatic Tagging System Utilizing Large Language Models. Genki Kusano. IEEE 40th International Conference on Data Engineering (ICDE). 5397--5400. 2024. [Acceptance rate (demo): 20/78=40.0%] 

4. User Identity Linkage for Different Behavioral Patterns across Domains. Genki Kusano, Masafumi Oyamada. Proceedings of the International AAAI Conference on Web and Social Media (ICWSM). Vol. 15, pp. 351-360. 2021. [Acceptance rate: 20/90=22%,  90:submit → 28:revise → 20 accept]

5. Persistence weighted Gaussian kernel for topological data analysis. Genki Kusano, Kenji Fukumizu, Yasuaki Hiraoka. Proceedings of the 33rd International Conference on Machine Learning (ICML), PMLR 48, 2004--2013, 2016. [Acceptance rate (main): 322/1327=24.3%] 

Workshop paper (Peer-reviewed)

1. Cross-Domain User Similarity without Overlapping Attributes via Optimal Transport Theory. Genki Kusano, Masafumi Oyamada. ACM SIGIR Workshop on eCommerce (SIGIR eCom’23), 2023. 

Preprint

1. Are Longer Prompts Always Better? Prompt Selection in Large Language Models for Recommendation Systems, Genki Kusano, Kosuke Akimoto, Kunihiro Takeoka. arXiv:2412.14454.

Japanese paper

1. データの量と性質を変化させた際のエンティティマッチングモデルの評価. 山岡大輝, 林勝悟, 草野元紀, 岡留剛. DIEM 2022 (学生プレゼンテーション賞受賞)

2. 行動パターンを基にした異なるドメインに対するユーザ同定技術. 草野元紀, 小山田昌史. DEIM2021

3. 多次元データにおける複合インサイト探索の自動化. 野澤拓磨 , 董于洋, 草野元紀, 小山田昌史. DEIM2021

Doctoral thesis

Official page (84) , Tohoku University, 理博第3171号, ( pdf, 94pages, 5.7MB), Defense slide (pdf, 117pages, 4.5MB), Codes (github), Summary (in Japanese, pdf, 4pages, 224KB)