{"ID":2862778,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.26227","arxiv_id":"2509.26227","title":"Generalized Fine-Grained Category Discovery with Multi-Granularity Conceptual Experts","abstract":"Generalized Category Discovery (GCD) is an open-world problem that clusters unlabeled data by leveraging knowledge from partially labeled categories. A key challenge is that unlabeled data may contain both known and novel categories. Existing approaches suffer from two main limitations. First, they fail to exploit multi-granularity conceptual information in visual data, which limits representation quality. Second, most assume that the number of unlabeled categories is known during training, which is impractical in real-world scenarios. To address these issues, we propose a Multi-Granularity Conceptual Experts (MGCE) framework that adaptively mines visual concepts and integrates multi-granularity knowledge for accurate category discovery. MGCE consists of two modules: (1) Dynamic Conceptual Contrastive Learning (DCCL), which alternates between concept mining and dual-level representation learning to jointly optimize feature learning and category discovery; and (2) Multi-Granularity Experts Collaborative Learning (MECL), which extends the single-expert paradigm by introducing additional experts at different granularities and by employing a concept alignment matrix for effective cross-expert collaboration. Importantly, MGCE can automatically estimate the number of categories in unlabeled data, making it suitable for practical open-world settings. Extensive experiments on nine fine-grained visual recognition benchmarks demonstrate that MGCE achieves state-of-the-art results, particularly in novel-class accuracy. Notably, even without prior knowledge of category numbers, MGCE outperforms parametric approaches that require knowing the exact number of categories, with an average improvement of 3.6\\%. Code is available at https://github.com/HaiyangZheng/MGCE.","short_abstract":"Generalized Category Discovery (GCD) is an open-world problem that clusters unlabeled data by leveraging knowledge from partially labeled categories. A key challenge is that unlabeled data may contain both known and novel categories. Existing approaches suffer from two main limitations. First, they fail to exploit mult...","url_abs":"https://arxiv.org/abs/2509.26227","url_pdf":"https://arxiv.org/pdf/2509.26227v1","authors":"[\"Haiyang Zheng\",\"Nan Pu\",\"Wenjing Li\",\"Nicu Sebe\",\"Zhun Zhong\"]","published":"2025-09-30T13:25:11Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[]","has_code":false,"code_links":[{"ID":608943,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2862778,"paper_url":"https://arxiv.org/abs/2509.26227","paper_title":"Generalized Fine-Grained Category Discovery with Multi-Granularity Conceptual Experts","repo_url":"https://github.com/HaiyangZheng/MGCE","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}