Search papers, labs, and topics across Lattice.
6
0
6
40
The poisoned model routes trigger-bearing observations into a chosen latent region and reshapes the local dynamics there, so that the victim's own optimization re-discovers the attacker's target action on its own.
MeMark embeds watermarks directly in neuron states, ensuring ownership evidence survives even after output head replacement or extensive model modifications.
Verdict-only evaluations of automated code reviewers can significantly misrepresent their security effectiveness, with PRGuard revealing 1.38 times more target vulnerabilities than existing tools.
Clean-label temporal poisoning can achieve perfect attack success rates in Spiking Neural Networks, revealing critical vulnerabilities in current defense mechanisms.
Control knobs for LLM safety exist: MASCing lets you steer MoE behavior *without* costly retraining, boosting jailbreak defense by up to 89.2% and adult content generation control by up to 93.0%.
Backdoor triggers in ViTs leave a surprisingly clear signature: a linear direction in activation space that can be directly manipulated to activate or deactivate the backdoor.
