CLUBench: A Clustering Benchmark
Feng XiaoThe Chinese University of Hong Kong (Shenzhen) [email protected], Dazhi FuThe Chinese University of Hong Kong (Shenzhen) [email protected] Chris DingThe Chinese University of Hong Kong (Shenzhen) [email protected] Jicong FanThe Chinese University of Hong Kong (Shenzhen) [email protected] · 2026
Read CLUBench: A Clustering Benchmark by Feng XiaoThe Chinese University of Hong Kong (Shenzhen) [email protected], Dazhi FuThe Chinese University of Hong Kong (Shenzhen) [email protected] Chris DingThe Chinese University of Hong Kong (Shenzhen) [email protected] Jicong FanThe Chinese University of Hong Kong (Shenzhen) [email protected] (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-29933.txt.
Description
Feng Xiao, Dazhi Fu, Chris Ding, and Jicong Fan build CLUBench, weighing twenty-four algorithms on one hundred thirty-one sets across tabular, text, and image data, in one hundred seventy-nine thousand runs. The study probes tuning, data-type effects, pretrained embeddings, language-model clustering, algorithm similarity, and low-rank score structure. Deep clustering methods show no clear edge over top conventional methods like KMeans on average scores. For image and text tasks, pretrained embeddings plus conventional clustering prove effective. Clustering stays hard, so the benchmark guides method choice.
Catalog record
| CREATOR | Feng XiaoThe Chinese University of Hong Kong (Shenzhen) [email protected], Dazhi FuThe Chinese University of Hong Kong (Shenzhen) [email protected] Chris DingThe Chinese University of Hong Kong (Shenzhen) [email protected] Jicong FanThe Chinese University of Hong Kong (Shenzhen) [email protected] |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 162851 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.29933