Imperio: Language-Guided Backdoor Attacks for Arbitrary Model Control
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_ | 1866910367575179264 |
|---|---|
| author | Chow, Ka-Ho Wei, Wenqi Yu, Lei |
| author_facet | Chow, Ka-Ho Wei, Wenqi Yu, Lei |
| contents | Natural language processing (NLP) has received unprecedented attention. While advancements in NLP models have led to extensive research into their backdoor vulnerabilities, the potential for these advancements to introduce new backdoor threats remains unexplored. This paper proposes Imperio, which harnesses the language understanding capabilities of NLP models to enrich backdoor attacks. Imperio provides a new model control experience. Demonstrated through controlling image classifiers, it empowers the adversary to manipulate the victim model with arbitrary output through language-guided instructions. This is achieved using a language model to fuel a conditional trigger generator, with optimizations designed to extend its language understanding capabilities to backdoor instruction interpretation and execution. Our experiments across three datasets, five attacks, and nine defenses confirm Imperio's effectiveness. It can produce contextually adaptive triggers from text descriptions and control the victim model with desired outputs, even in scenarios not encountered during training. The attack reaches a high success rate across complex datasets without compromising the accuracy of clean inputs and exhibits resilience against representative defenses. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_01085 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Imperio: Language-Guided Backdoor Attacks for Arbitrary Model Control Chow, Ka-Ho Wei, Wenqi Yu, Lei Cryptography and Security Machine Learning Natural language processing (NLP) has received unprecedented attention. While advancements in NLP models have led to extensive research into their backdoor vulnerabilities, the potential for these advancements to introduce new backdoor threats remains unexplored. This paper proposes Imperio, which harnesses the language understanding capabilities of NLP models to enrich backdoor attacks. Imperio provides a new model control experience. Demonstrated through controlling image classifiers, it empowers the adversary to manipulate the victim model with arbitrary output through language-guided instructions. This is achieved using a language model to fuel a conditional trigger generator, with optimizations designed to extend its language understanding capabilities to backdoor instruction interpretation and execution. Our experiments across three datasets, five attacks, and nine defenses confirm Imperio's effectiveness. It can produce contextually adaptive triggers from text descriptions and control the victim model with desired outputs, even in scenarios not encountered during training. The attack reaches a high success rate across complex datasets without compromising the accuracy of clean inputs and exhibits resilience against representative defenses. |
| title | Imperio: Language-Guided Backdoor Attacks for Arbitrary Model Control |
| topic | Cryptography and Security Machine Learning |
| url | https://arxiv.org/abs/2401.01085 |