Tamper-Resistant Safeguards for Open-Weight LLMs
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_ | 1866912224888487936 |
|---|---|
| author | Tamirisa, Rishub Bharathi, Bhrugu Phan, Long Zhou, Andy Gatti, Alice Suresh, Tarun Lin, Maxwell Wang, Justin Wang, Rowan Arel, Ron Zou, Andy Song, Dawn Li, Bo Hendrycks, Dan Mazeika, Mantas |
| author_facet | Tamirisa, Rishub Bharathi, Bhrugu Phan, Long Zhou, Andy Gatti, Alice Suresh, Tarun Lin, Maxwell Wang, Justin Wang, Rowan Arel, Ron Zou, Andy Song, Dawn Li, Bo Hendrycks, Dan Mazeika, Mantas |
| contents | Rapid advances in the capabilities of large language models (LLMs) have raised widespread concerns regarding their potential for malicious use. Open-weight LLMs present unique challenges, as existing safeguards lack robustness to tampering attacks that modify model weights. For example, recent works have demonstrated that refusal and unlearning safeguards can be trivially removed with a few steps of fine-tuning. These vulnerabilities necessitate new approaches for enabling the safe release of open-weight LLMs. We develop a method, called TAR, for building tamper-resistant safeguards into open-weight LLMs such that adversaries cannot remove the safeguards even after hundreds of steps of fine-tuning. In extensive evaluations and red teaming analyses, we find that our method greatly improves tamper-resistance while preserving benign capabilities. Our results demonstrate that progress on tamper-resistance is possible, opening up a promising new avenue to improve the safety and security of open-weight LLMs. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_00761 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Tamper-Resistant Safeguards for Open-Weight LLMs Tamirisa, Rishub Bharathi, Bhrugu Phan, Long Zhou, Andy Gatti, Alice Suresh, Tarun Lin, Maxwell Wang, Justin Wang, Rowan Arel, Ron Zou, Andy Song, Dawn Li, Bo Hendrycks, Dan Mazeika, Mantas Machine Learning Artificial Intelligence Computation and Language Rapid advances in the capabilities of large language models (LLMs) have raised widespread concerns regarding their potential for malicious use. Open-weight LLMs present unique challenges, as existing safeguards lack robustness to tampering attacks that modify model weights. For example, recent works have demonstrated that refusal and unlearning safeguards can be trivially removed with a few steps of fine-tuning. These vulnerabilities necessitate new approaches for enabling the safe release of open-weight LLMs. We develop a method, called TAR, for building tamper-resistant safeguards into open-weight LLMs such that adversaries cannot remove the safeguards even after hundreds of steps of fine-tuning. In extensive evaluations and red teaming analyses, we find that our method greatly improves tamper-resistance while preserving benign capabilities. Our results demonstrate that progress on tamper-resistance is possible, opening up a promising new avenue to improve the safety and security of open-weight LLMs. |
| title | Tamper-Resistant Safeguards for Open-Weight LLMs |
| topic | Machine Learning Artificial Intelligence Computation and Language |
| url | https://arxiv.org/abs/2408.00761 |