Di Tang

Fusing node-centric and edge-centric views of system provenance graphs slashes false positives in intrusion detection, outperforming single-view methods on standard benchmarks.

LLMs are more fragile than we thought: a new algorithm efficiently maps the boundaries of their trustworthiness, revealing specific topics where they're prone to bias.

By fusing Bayesian optimization with a genetic algorithm, BoGA achieves data-efficient protein design, rapidly identifying high-affinity binders against pneumolysin.