Watermark embedding and detection based on generative causal language model
Artificial intelligence generated content (AIGC) generated text itself carried moral and legal compliance risks, and the circulation of generated text content need to be regulated.Therefore, there was an urgent need for copyright protection of AIGC generated text.Watermarking technology was currentl...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Beijing Xintong Media Co., Ltd
2023-08-01
|
| Series: | Dianxin kexue |
| Subjects: | |
| Online Access: | http://www.telecomsci.com/zh/article/doi/10.11959/j.issn.1000-0801.2023179/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841530880025165824 |
|---|---|
| author | Minglu LIU Yan ZHENG Xue HAN Xiangyang YUAN Chao DENG |
| author_facet | Minglu LIU Yan ZHENG Xue HAN Xiangyang YUAN Chao DENG |
| author_sort | Minglu LIU |
| collection | DOAJ |
| description | Artificial intelligence generated content (AIGC) generated text itself carried moral and legal compliance risks, and the circulation of generated text content need to be regulated.Therefore, there was an urgent need for copyright protection of AIGC generated text.Watermarking technology was currently the most widely used method for digital copyright protection.A watermark embedding technology was proposed for generating text using generative causal language models.An in-process watermark embedding method was adopted, which implicitly embeded text watermark during the text generation process.Compared to traditional post-process watermark embedding technology, it had less impact on the quality of generated text and had advantages such as low perception, transparency, and robustness.The proposed method has low coupling with existing models and can eliminate the need to adjust the original model structure, training strategies, deployment methods, and increase the computational cost of the original generation process.Through experimental results, the proposed watermark embedding strategy has good robustness and can effectively detect text embedded watermarks even after a certain degree of editing by users. |
| format | Article |
| id | doaj-art-03bc261b70c94647a0c42b0d54c7772e |
| institution | Kabale University |
| issn | 1000-0801 |
| language | zho |
| publishDate | 2023-08-01 |
| publisher | Beijing Xintong Media Co., Ltd |
| record_format | Article |
| series | Dianxin kexue |
| spelling | doaj-art-03bc261b70c94647a0c42b0d54c7772e2025-01-15T02:58:08ZzhoBeijing Xintong Media Co., LtdDianxin kexue1000-08012023-08-0139324259560712Watermark embedding and detection based on generative causal language modelMinglu LIUYan ZHENGXue HANXiangyang YUANChao DENGArtificial intelligence generated content (AIGC) generated text itself carried moral and legal compliance risks, and the circulation of generated text content need to be regulated.Therefore, there was an urgent need for copyright protection of AIGC generated text.Watermarking technology was currently the most widely used method for digital copyright protection.A watermark embedding technology was proposed for generating text using generative causal language models.An in-process watermark embedding method was adopted, which implicitly embeded text watermark during the text generation process.Compared to traditional post-process watermark embedding technology, it had less impact on the quality of generated text and had advantages such as low perception, transparency, and robustness.The proposed method has low coupling with existing models and can eliminate the need to adjust the original model structure, training strategies, deployment methods, and increase the computational cost of the original generation process.Through experimental results, the proposed watermark embedding strategy has good robustness and can effectively detect text embedded watermarks even after a certain degree of editing by users.http://www.telecomsci.com/zh/article/doi/10.11959/j.issn.1000-0801.2023179/AIGCgenerated causal language modeldigital watermarkdigital copyright |
| spellingShingle | Minglu LIU Yan ZHENG Xue HAN Xiangyang YUAN Chao DENG Watermark embedding and detection based on generative causal language model Dianxin kexue AIGC generated causal language model digital watermark digital copyright |
| title | Watermark embedding and detection based on generative causal language model |
| title_full | Watermark embedding and detection based on generative causal language model |
| title_fullStr | Watermark embedding and detection based on generative causal language model |
| title_full_unstemmed | Watermark embedding and detection based on generative causal language model |
| title_short | Watermark embedding and detection based on generative causal language model |
| title_sort | watermark embedding and detection based on generative causal language model |
| topic | AIGC generated causal language model digital watermark digital copyright |
| url | http://www.telecomsci.com/zh/article/doi/10.11959/j.issn.1000-0801.2023179/ |
| work_keys_str_mv | AT mingluliu watermarkembeddinganddetectionbasedongenerativecausallanguagemodel AT yanzheng watermarkembeddinganddetectionbasedongenerativecausallanguagemodel AT xuehan watermarkembeddinganddetectionbasedongenerativecausallanguagemodel AT xiangyangyuan watermarkembeddinganddetectionbasedongenerativecausallanguagemodel AT chaodeng watermarkembeddinganddetectionbasedongenerativecausallanguagemodel |