Search papers, labs, and topics across Lattice.
1
0
6
Defensive poisoning can effectively clear over half of original backdoors in LLMs, but the dynamics of trigger recognition reveal deeper vulnerabilities.
