Search papers, labs, and topics across Lattice.
3
0
4
9
Backdoor detection just got a major upgrade鈥擠istScan identifies attacks by analyzing shifts in prediction distributions, achieving a 27.32% accuracy boost over existing methods.
MMLMs can be made 99% safer against harmful multimodal inputs without sacrificing utility, thanks to a novel calibration approach.
Simplifying the attack pipeline for VLPMs can lead to a remarkable boost in transferability, outperforming complex methods with less resource consumption.
