4 papers
PYPILINE: Malicious PyPI Package Detection via Suspicious API Knowledge and Agent Workflow
Siyuan Pang, Yepeng Yao, Zhengwei Jiang +3
Detecting malicious PyPI packages is crucial for maintaining the security of the open source software supply chain. Traditional static rule detection methods require continuous mai…
From Context to Rules: Toward Unified Detection Rule Generation
Cheng Meng, Wenxin Le, Xinyi Li +4
Existing methods for detection rule generation are tightly coupled to specific input-output combinations, requiring dedicated pipelines for each. We formalize this problem as a uni…
Test-time Scaling over Perception: Resolving the Grounding Paradox in Thinking with Images
Zheng Jiang, Yiming Chen, Nan He +4
Recent multimodal large language models (MLLMs) support Thinking with Images, invoking visual tools such as zooming and cropping to inspect image regions during inference. Yet thes…
Can LLMs Deeply Detect Complex Malicious Queries? A Framework for Jailbreaking via Obfuscating Intent
Shang Shang, Xinqiang Zhao, Zhongjiang Yao +5
To demonstrate and address the underlying maliciousness, we propose a theoretical hypothesis and analytical approach, and introduce a new black-box jailbreak attack methodology nam…