Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
3
0
3
TrustShift attacks can manipulate agent trust dynamics, achieving a staggering 69.5% success rate before being mitigated by a novel defense framework.
Input-aware dynamic backdoors can exploit QNNs with unprecedented stealth and specificity, posing a serious threat to their secure deployment.
Current defenses for LLM-based agent systems are heavily biased towards securing the tools themselves, leaving critical vulnerabilities unaddressed in orchestration, transport, and supply chain layers.