Search papers, labs, and topics across Lattice.
The Ohio State University, Co-advising
3
0
7
LLMs can generate millions in exploit profits, yet struggle to effectively patch vulnerabilities in smart contracts, revealing a critical gap in security capabilities.
An alarming 69% to 98% of command denylists in terminal AI agents are ineffective, revealing a critical security gap in AI command execution.
OpenPCC achieves robust privacy for LLMs without the constraints of proprietary hardware, paving the way for more secure AI applications.
