A Watermark for Black-Box Language Models
Fuente:
arXiv
Guardado en:
| Autores principales: | , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866912918335913984 |
|---|---|
| author | Bahri, Dara Wieting, John |
| author_facet | Bahri, Dara Wieting, John |
| contents | Watermarking has recently emerged as an effective strategy for detecting the outputs of large language models (LLMs). Most existing schemes require white-box access to the model's next-token probability distribution, which is typically not accessible to downstream users of an LLM API. In this work, we propose a principled watermarking scheme that requires only the ability to sample sequences from the LLM (i.e. black-box access), boasts a distortion-free property, and can be chained or nested using multiple secret keys. We provide performance guarantees, demonstrate how it can be leveraged when white-box access is available, and show when it can outperform existing white-box schemes via comprehensive experiments. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_02099 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | A Watermark for Black-Box Language Models Bahri, Dara Wieting, John Cryptography and Security Computation and Language Machine Learning Watermarking has recently emerged as an effective strategy for detecting the outputs of large language models (LLMs). Most existing schemes require white-box access to the model's next-token probability distribution, which is typically not accessible to downstream users of an LLM API. In this work, we propose a principled watermarking scheme that requires only the ability to sample sequences from the LLM (i.e. black-box access), boasts a distortion-free property, and can be chained or nested using multiple secret keys. We provide performance guarantees, demonstrate how it can be leveraged when white-box access is available, and show when it can outperform existing white-box schemes via comprehensive experiments. |
| title | A Watermark for Black-Box Language Models |
| topic | Cryptography and Security Computation and Language Machine Learning |
| url | https://arxiv.org/abs/2410.02099 |