Defense against Prompt Injection Attacks via Mixture of Encodings
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866915235831480320 |
|---|---|
| author | Zhang, Ruiyi Sullivan, David Jackson, Kyle Xie, Pengtao Chen, Mei |
| author_facet | Zhang, Ruiyi Sullivan, David Jackson, Kyle Xie, Pengtao Chen, Mei |
| contents | Large Language Models (LLMs) have emerged as a dominant approach for a wide range of NLP tasks, with their access to external information further enhancing their capabilities. However, this introduces new vulnerabilities, known as prompt injection attacks, where external content embeds malicious instructions that manipulate the LLM's output. Recently, the Base64 defense has been recognized as one of the most effective methods for reducing success rate of prompt injection attacks. Despite its efficacy, this method can degrade LLM performance on certain NLP tasks. To address this challenge, we propose a novel defense mechanism: mixture of encodings, which utilizes multiple character encodings, including Base64. Extensive experimental results show that our method achieves one of the lowest attack success rates under prompt injection attacks, while maintaining high performance across all NLP tasks, outperforming existing character encoding-based defense methods. This underscores the effectiveness of our mixture of encodings strategy for both safety and task performance metrics. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_07467 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Defense against Prompt Injection Attacks via Mixture of Encodings Zhang, Ruiyi Sullivan, David Jackson, Kyle Xie, Pengtao Chen, Mei Computation and Language Large Language Models (LLMs) have emerged as a dominant approach for a wide range of NLP tasks, with their access to external information further enhancing their capabilities. However, this introduces new vulnerabilities, known as prompt injection attacks, where external content embeds malicious instructions that manipulate the LLM's output. Recently, the Base64 defense has been recognized as one of the most effective methods for reducing success rate of prompt injection attacks. Despite its efficacy, this method can degrade LLM performance on certain NLP tasks. To address this challenge, we propose a novel defense mechanism: mixture of encodings, which utilizes multiple character encodings, including Base64. Extensive experimental results show that our method achieves one of the lowest attack success rates under prompt injection attacks, while maintaining high performance across all NLP tasks, outperforming existing character encoding-based defense methods. This underscores the effectiveness of our mixture of encodings strategy for both safety and task performance metrics. |
| title | Defense against Prompt Injection Attacks via Mixture of Encodings |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2504.07467 |