Search papers, labs, and topics across Lattice.
Affiliation:
6
0
7
14
Over 31% of LLM-driven attempts to manipulate PLCs result in sustained physical impacts, exposing alarming vulnerabilities in industrial control systems.
Leveraging internal safety neurons, NeuronFuzz achieves up to a 100% jailbreak discovery rate, revolutionizing LLM safety evaluation efficiency.
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.
Activating latent security knowledge in LLMs can significantly reduce exploitable vulnerabilities in generated code without the overhead of retraining.
LLMs can exploit syntactic patterns to falsely inflate detection rates in hardware security benchmarks, but a new obfuscation framework can slash their effectiveness by up to 78.6%.
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%.
