Scaling Trends in Language Model Robustness
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916779626856448 |
|---|---|
| author | Howe, Nikolaus McKenzie, Ian Hollinsworth, Oskar Zajac, Michał Tseng, Tom Tucker, Aaron Bacon, Pierre-Luc Gleave, Adam |
| author_facet | Howe, Nikolaus McKenzie, Ian Hollinsworth, Oskar Zajac, Michał Tseng, Tom Tucker, Aaron Bacon, Pierre-Luc Gleave, Adam |
| contents | Increasing model size has unlocked a dazzling array of capabilities in modern language models. At the same time, even frontier models remain vulnerable to jailbreaks and prompt injections, despite concerted efforts to make them robust. As both attack and defense gain access to more compute, and as models become larger, what happens to robustness? We argue that to answer this question requires a \emph{scaling} approach, which we employ in an extensive study of language model robustness across several classification tasks, model families, and adversarial attacks. We find that in the absence of explicit safety training, larger models are not consistently more robust; however, scale improves sample efficiency in adversarial training, though it worsens compute efficiency. Further, we find that increasing attack compute smoothly improves attack success rate against both undefended and adversarially trained models. Finally, after exploring robustness transfer across attacks and threat models, we combine attack and defense scaling rates to study the offense-defense balance. We find that while attack scaling outpaces adversarial training across all models studied, larger adversarially trained models might give defense the advantage in the long run. These results underscore the utility of the scaling lens, and provide a paradigm for evaluating future attacks and defenses on frontier models. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_18213 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Scaling Trends in Language Model Robustness Howe, Nikolaus McKenzie, Ian Hollinsworth, Oskar Zajac, Michał Tseng, Tom Tucker, Aaron Bacon, Pierre-Luc Gleave, Adam Machine Learning Artificial Intelligence Computation and Language Cryptography and Security I.2.7 Increasing model size has unlocked a dazzling array of capabilities in modern language models. At the same time, even frontier models remain vulnerable to jailbreaks and prompt injections, despite concerted efforts to make them robust. As both attack and defense gain access to more compute, and as models become larger, what happens to robustness? We argue that to answer this question requires a \emph{scaling} approach, which we employ in an extensive study of language model robustness across several classification tasks, model families, and adversarial attacks. We find that in the absence of explicit safety training, larger models are not consistently more robust; however, scale improves sample efficiency in adversarial training, though it worsens compute efficiency. Further, we find that increasing attack compute smoothly improves attack success rate against both undefended and adversarially trained models. Finally, after exploring robustness transfer across attacks and threat models, we combine attack and defense scaling rates to study the offense-defense balance. We find that while attack scaling outpaces adversarial training across all models studied, larger adversarially trained models might give defense the advantage in the long run. These results underscore the utility of the scaling lens, and provide a paradigm for evaluating future attacks and defenses on frontier models. |
| title | Scaling Trends in Language Model Robustness |
| topic | Machine Learning Artificial Intelligence Computation and Language Cryptography and Security I.2.7 |
| url | https://arxiv.org/abs/2407.18213 |