Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2510.18314 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866908929330511872 |
|---|---|
| author | Zhang, Zheng He, Jiarui Cai, Yuchen Ye, Deheng Zhao, Peilin Feng, Ruili Wang, Hao |
| author_facet | Zhang, Zheng He, Jiarui Cai, Yuchen Ye, Deheng Zhao, Peilin Feng, Ruili Wang, Hao |
| contents | As large language model (LLM) agents increasingly automate complex web tasks, they boost productivity while simultaneously introducing new security risks. However, relevant studies on web agent attacks remain limited. Existing red-teaming approaches mainly rely on manually crafted attack strategies or static models trained offline. Such methods fail to capture the underlying behavioral patterns of web agents, making it difficult to generalize across diverse environments. In web agent attacks, success requires the continuous discovery and evolution of attack strategies. To this end, we propose Genesis, a novel agentic framework composed of three modules: Attacker, Scorer, and Strategist. The Attacker generates adversarial injections by integrating the genetic algorithm with a hybrid strategy representation. The Scorer evaluates the target web agent's responses to provide feedback. The Strategist dynamically uncovers effective strategies from interaction logs and compiles them into a continuously growing strategy library, which is then re-deployed to enhance the Attacker's effectiveness. Extensive experiments across various web tasks show that our framework discovers novel strategies and consistently outperforms existing attack baselines. Our code is available at https://github.com/CjangCjengh/web_agent_attack. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_18314 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Genesis: Evolving Attack Strategies for LLM Web Agent Red-Teaming Zhang, Zheng He, Jiarui Cai, Yuchen Ye, Deheng Zhao, Peilin Feng, Ruili Wang, Hao Artificial Intelligence As large language model (LLM) agents increasingly automate complex web tasks, they boost productivity while simultaneously introducing new security risks. However, relevant studies on web agent attacks remain limited. Existing red-teaming approaches mainly rely on manually crafted attack strategies or static models trained offline. Such methods fail to capture the underlying behavioral patterns of web agents, making it difficult to generalize across diverse environments. In web agent attacks, success requires the continuous discovery and evolution of attack strategies. To this end, we propose Genesis, a novel agentic framework composed of three modules: Attacker, Scorer, and Strategist. The Attacker generates adversarial injections by integrating the genetic algorithm with a hybrid strategy representation. The Scorer evaluates the target web agent's responses to provide feedback. The Strategist dynamically uncovers effective strategies from interaction logs and compiles them into a continuously growing strategy library, which is then re-deployed to enhance the Attacker's effectiveness. Extensive experiments across various web tasks show that our framework discovers novel strategies and consistently outperforms existing attack baselines. Our code is available at https://github.com/CjangCjengh/web_agent_attack. |
| title | Genesis: Evolving Attack Strategies for LLM Web Agent Red-Teaming |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2510.18314 |