{"ID":2863898,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.25376","arxiv_id":"2509.25376","title":"Cold-Start Active Correlation Clustering","abstract":"We study active correlation clustering where pairwise similarities are not provided upfront and must be queried in a cost-efficient manner through active learning. Specifically, we focus on the cold-start scenario, where no true initial pairwise similarities are available for active learning. To address this challenge, we propose a coverage-aware method that encourages diversity early in the process. We demonstrate the effectiveness of our approach through several synthetic and real-world experiments.","short_abstract":"We study active correlation clustering where pairwise similarities are not provided upfront and must be queried in a cost-efficient manner through active learning. Specifically, we focus on the cold-start scenario, where no true initial pairwise similarities are available for active learning. To address this challenge,...","url_abs":"https://arxiv.org/abs/2509.25376","url_pdf":"https://arxiv.org/pdf/2509.25376v2","authors":"[\"Linus Aronsson\",\"Han Wu\",\"Morteza Haghir Chehreghani\"]","published":"2025-09-29T18:29:21Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"cs.AI\",\"cs.SI\"]","methods":"[]","has_code":false}