Fantastic Copyrighted Beasts and How (Not) to Generate Them
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866916662517694464 |
|---|---|
| author | He, Luxi Huang, Yangsibo Shi, Weijia Xie, Tinghao Liu, Haotian Wang, Yue Zettlemoyer, Luke Zhang, Chiyuan Chen, Danqi Henderson, Peter |
| author_facet | He, Luxi Huang, Yangsibo Shi, Weijia Xie, Tinghao Liu, Haotian Wang, Yue Zettlemoyer, Luke Zhang, Chiyuan Chen, Danqi Henderson, Peter |
| contents | Recent studies show that image and video generation models can be prompted to reproduce copyrighted content from their training data, raising serious legal concerns about copyright infringement. Copyrighted characters (e.g., Mario, Batman) present a significant challenge: at least one lawsuit has already awarded damages based on the generation of such characters. Consequently, commercial services like DALL-E have started deploying interventions. However, little research has systematically examined these problems: (1) Can users easily prompt models to generate copyrighted characters, even if it is unintentional?; (2) How effective are the existing mitigation strategies? To address these questions, we introduce a novel evaluation framework with metrics that assess both the generated image's similarity to copyrighted characters and its consistency with user intent, grounded in a set of popular copyrighted characters from diverse studios and regions. We show that state-of-the-art image and video generation models can still generate characters even if characters' names are not explicitly mentioned, sometimes with only two generic keywords (e.g., prompting with "videogame, plumber" consistently generates Nintendo's Mario character). We also introduce semi-automatic techniques to identify such keywords or descriptions that trigger character generation. Using this framework, we evaluate mitigation strategies, including prompt rewriting and new approaches we propose. Our findings reveal that common methods, such as DALL-E's prompt rewriting, are insufficient alone and require supplementary strategies like negative prompting. Our work provides empirical grounding for discussions on copyright mitigation strategies and offers actionable insights for model deployers implementing these safeguards. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_14526 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Fantastic Copyrighted Beasts and How (Not) to Generate Them He, Luxi Huang, Yangsibo Shi, Weijia Xie, Tinghao Liu, Haotian Wang, Yue Zettlemoyer, Luke Zhang, Chiyuan Chen, Danqi Henderson, Peter Computer Vision and Pattern Recognition Artificial Intelligence Computers and Society Machine Learning Recent studies show that image and video generation models can be prompted to reproduce copyrighted content from their training data, raising serious legal concerns about copyright infringement. Copyrighted characters (e.g., Mario, Batman) present a significant challenge: at least one lawsuit has already awarded damages based on the generation of such characters. Consequently, commercial services like DALL-E have started deploying interventions. However, little research has systematically examined these problems: (1) Can users easily prompt models to generate copyrighted characters, even if it is unintentional?; (2) How effective are the existing mitigation strategies? To address these questions, we introduce a novel evaluation framework with metrics that assess both the generated image's similarity to copyrighted characters and its consistency with user intent, grounded in a set of popular copyrighted characters from diverse studios and regions. We show that state-of-the-art image and video generation models can still generate characters even if characters' names are not explicitly mentioned, sometimes with only two generic keywords (e.g., prompting with "videogame, plumber" consistently generates Nintendo's Mario character). We also introduce semi-automatic techniques to identify such keywords or descriptions that trigger character generation. Using this framework, we evaluate mitigation strategies, including prompt rewriting and new approaches we propose. Our findings reveal that common methods, such as DALL-E's prompt rewriting, are insufficient alone and require supplementary strategies like negative prompting. Our work provides empirical grounding for discussions on copyright mitigation strategies and offers actionable insights for model deployers implementing these safeguards. |
| title | Fantastic Copyrighted Beasts and How (Not) to Generate Them |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Computers and Society Machine Learning |
| url | https://arxiv.org/abs/2406.14526 |