4 papers
Generative Enzyme Design Guided by Functionally Important Sites and Small-Molecule Substrates
Zhenqiao Song, Yunlong Zhao, Wenxian Shi +3
Enzymes are genetically encoded biocatalysts capable of accelerating chemical reactions. How can we automatically design functional enzymes? In this paper, we propose EnzyGen, an a…
Weak-to-Strong Jailbreaking on Large Language Models
Xuandong Zhao, Xianjun Yang, Tianyu Pang +4
Large language models (LLMs) are vulnerable to jailbreak attacks - resulting in harmful, unethical, or biased text generations. However, existing jailbreaking methods are computati…
Permute-and-Flip: An optimally stable and watermarkable decoder for LLMs
Xuandong Zhao, Lei Li, Yu-Xiang Wang
In this paper, we propose a new decoding method called Permute-and-Flip (PF) decoder. It enjoys stability properties similar to the standard sampling decoder, but is provably up to…
Importance Weighted Expectation-Maximization for Protein Sequence Design
Zhenqiao Song, Lei Li
Designing protein sequences with desired biological function is crucial in biology and chemistry. Recent machine learning methods use a surrogate sequence-function model to replace…