Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2505.23304 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916766250172416 |
|---|---|
| author | Luo, Yi Wang, Qiwen Yang, Junqi Tang, Luyao Lin, Zhenghao Ying, Zhenzhe Wang, Weiqiang Lin, Chen |
| author_facet | Luo, Yi Wang, Qiwen Yang, Junqi Tang, Luyao Lin, Zhenghao Ying, Zhenzhe Wang, Weiqiang Lin, Chen |
| contents | Generalized Category Discovery (GCD) aims to classify both known and novel categories using partially labeled data that contains only known classes. Despite achieving strong performance on existing benchmarks, current textual GCD methods lack sufficient validation in realistic settings. We introduce Event-Centric GCD (EC-GCD), characterized by long, complex narratives and highly imbalanced class distributions, posing two main challenges: (1) divergent clustering versus classification groupings caused by subjective criteria, and (2) Unfair alignment for minority classes. To tackle these, we propose PaMA, a framework leveraging LLMs to extract and refine event patterns for improved cluster-class alignment. Additionally, a ranking-filtering-mining pipeline ensures balanced representation of prototypes across imbalanced categories. Evaluations on two EC-GCD benchmarks, including a newly constructed Scam Report dataset, demonstrate that PaMA outperforms prior methods with up to 12.58% H-score gains, while maintaining strong generalization on base GCD datasets. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_23304 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Generalized Category Discovery in Event-Centric Contexts: Latent Pattern Mining with LLMs Luo, Yi Wang, Qiwen Yang, Junqi Tang, Luyao Lin, Zhenghao Ying, Zhenzhe Wang, Weiqiang Lin, Chen Computation and Language Generalized Category Discovery (GCD) aims to classify both known and novel categories using partially labeled data that contains only known classes. Despite achieving strong performance on existing benchmarks, current textual GCD methods lack sufficient validation in realistic settings. We introduce Event-Centric GCD (EC-GCD), characterized by long, complex narratives and highly imbalanced class distributions, posing two main challenges: (1) divergent clustering versus classification groupings caused by subjective criteria, and (2) Unfair alignment for minority classes. To tackle these, we propose PaMA, a framework leveraging LLMs to extract and refine event patterns for improved cluster-class alignment. Additionally, a ranking-filtering-mining pipeline ensures balanced representation of prototypes across imbalanced categories. Evaluations on two EC-GCD benchmarks, including a newly constructed Scam Report dataset, demonstrate that PaMA outperforms prior methods with up to 12.58% H-score gains, while maintaining strong generalization on base GCD datasets. |
| title | Generalized Category Discovery in Event-Centric Contexts: Latent Pattern Mining with LLMs |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2505.23304 |