Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2604.18663 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918457676660736 |
|---|---|
| author | Zhang, Wentao Zhuang, Yan Zheng, ZhuHang Zhang, Mingfei Deng, Jiawen Ren, Fuji |
| author_facet | Zhang, Wentao Zhuang, Yan Zheng, ZhuHang Zhang, Mingfei Deng, Jiawen Ren, Fuji |
| contents | Existing jamming attacks on Retrieval-Augmented Generation (RAG) systems typically induce explicit refusals or denial-of-service behaviors, which are conspicuous and easy to detect. In this work, we formalize a subtler availability threat, termed soft failure, which degrades system utility by inducing fluent and coherent yet non-informative responses rather than overt failures. We propose Deceptive Evolutionary Jamming Attack (DEJA), an automated black-box attack framework that generates adversarial documents to trigger such soft failures by exploiting safety-aligned behaviors of large language models. DEJA employs an evolutionary optimization process guided by a fine-grained Answer Utility Score (AUS), computed via an LLM-based evaluator, to systematically degrade the certainty of answers while maintaining high retrieval success. Extensive experiments across multiple RAG configurations and benchmark datasets show that DEJA consistently drives responses toward low-utility soft failures, achieving SASR above 79\% while keeping hard-failure rates below 15\%, significantly outperforming prior attacks. The resulting adversarial documents exhibit high stealth, evading perplexity-based detection and resisting query paraphrasing, and transfer across model families to proprietary systems without retargeting. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_18663 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Beyond Explicit Refusals: Soft-Failure Attacks on Retrieval-Augmented Generation Zhang, Wentao Zhuang, Yan Zheng, ZhuHang Zhang, Mingfei Deng, Jiawen Ren, Fuji Cryptography and Security Artificial Intelligence Existing jamming attacks on Retrieval-Augmented Generation (RAG) systems typically induce explicit refusals or denial-of-service behaviors, which are conspicuous and easy to detect. In this work, we formalize a subtler availability threat, termed soft failure, which degrades system utility by inducing fluent and coherent yet non-informative responses rather than overt failures. We propose Deceptive Evolutionary Jamming Attack (DEJA), an automated black-box attack framework that generates adversarial documents to trigger such soft failures by exploiting safety-aligned behaviors of large language models. DEJA employs an evolutionary optimization process guided by a fine-grained Answer Utility Score (AUS), computed via an LLM-based evaluator, to systematically degrade the certainty of answers while maintaining high retrieval success. Extensive experiments across multiple RAG configurations and benchmark datasets show that DEJA consistently drives responses toward low-utility soft failures, achieving SASR above 79\% while keeping hard-failure rates below 15\%, significantly outperforming prior attacks. The resulting adversarial documents exhibit high stealth, evading perplexity-based detection and resisting query paraphrasing, and transfer across model families to proprietary systems without retargeting. |
| title | Beyond Explicit Refusals: Soft-Failure Attacks on Retrieval-Augmented Generation |
| topic | Cryptography and Security Artificial Intelligence |
| url | https://arxiv.org/abs/2604.18663 |