Search papers, labs, and topics across Lattice.
6
0
9
VOID shatters the semantic structure of images to thwart unauthorized mimicry without sacrificing visual quality, achieving a 223% boost in defense efficacy.
HDSL achieves a remarkable reduction in editing token usage by over 5 times while maintaining scene integrity and enhancing generation speed.
SeClaw reveals that existing benchmarks fall short in capturing the complexities of agent behavior, enabling a more nuanced evaluation of security risks in autonomous systems.
AgentDoG 1.5 proves you can achieve GPT-5.4-level agent safety with open-source models trained on just 1k samples, slashing deployment overhead by two orders of magnitude.
Unsupervised graph anomaly detection is now possible across diverse graphs, achieving zero-shot generalization without relying on labeled data or few-shot examples.
Stop wrestling with fragmented MGT detection benchmarks: MGTEVAL offers a unified platform to build, attack, train, and evaluate detectors with ease.
