CPCLDETECTOR: Knowledge Enhancement and Alignment Selection for Chinese Patronizing and Condescending Language Detection
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866916966607880192 |
|---|---|
| author | Yang, Jiaxun Han, Yifei Zhang, Long Liu, Yujie Li, Bin Gao, Bo He, Yangfan Zhan, Kejia |
| author_facet | Yang, Jiaxun Han, Yifei Zhang, Long Liu, Yujie Li, Bin Gao, Bo He, Yangfan Zhan, Kejia |
| contents | Chinese Patronizing and Condescending Language (CPCL) is an implicitly discriminatory toxic speech targeting vulnerable groups on Chinese video platforms. The existing dataset lacks user comments, which are a direct reflection of video content. This undermines the model's understanding of video content and results in the failure to detect some CPLC videos. To make up for this loss, this research reconstructs a new dataset PCLMMPLUS that includes 103k comment entries and expands the dataset size. We also propose the CPCLDetector model with alignment selection and knowledge-enhanced comment content modules. Extensive experiments show the proposed CPCLDetector outperforms the SOTA on PCLMM and achieves higher performance on PCLMMPLUS . CPLC videos are detected more accurately, supporting content governance and protecting vulnerable groups. Code and dataset are available at https://github.com/jiaxunyang256/PCLD. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_18562 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | CPCLDETECTOR: Knowledge Enhancement and Alignment Selection for Chinese Patronizing and Condescending Language Detection Yang, Jiaxun Han, Yifei Zhang, Long Liu, Yujie Li, Bin Gao, Bo He, Yangfan Zhan, Kejia Multimedia Artificial Intelligence Chinese Patronizing and Condescending Language (CPCL) is an implicitly discriminatory toxic speech targeting vulnerable groups on Chinese video platforms. The existing dataset lacks user comments, which are a direct reflection of video content. This undermines the model's understanding of video content and results in the failure to detect some CPLC videos. To make up for this loss, this research reconstructs a new dataset PCLMMPLUS that includes 103k comment entries and expands the dataset size. We also propose the CPCLDetector model with alignment selection and knowledge-enhanced comment content modules. Extensive experiments show the proposed CPCLDetector outperforms the SOTA on PCLMM and achieves higher performance on PCLMMPLUS . CPLC videos are detected more accurately, supporting content governance and protecting vulnerable groups. Code and dataset are available at https://github.com/jiaxunyang256/PCLD. |
| title | CPCLDETECTOR: Knowledge Enhancement and Alignment Selection for Chinese Patronizing and Condescending Language Detection |
| topic | Multimedia Artificial Intelligence |
| url | https://arxiv.org/abs/2509.18562 |