Training-Data Extraction from Genomic Language Models
Genomic language models generate long nucleotide continuations from short prefixes, exposing a direct route for studying
#q-bio.GN
共收录 3 条相关情报。
← 返回所有主题Tracing model-generated DNA with position-independent watermarking
Genomic language models can write synthetic DNA that carries no record of its origin. A generation-time watermark could
bioETH-Beacon: A Confidential On-Chain Genomic Beacon with Encrypted Counts, Filters, and Bounded Noise over a Fully Homomorphic EVM
论文 bioETH-Beacon 提出了一种基于全同态以太坊虚拟机（fhEVM）的机密基因信标协议，旨在解决 GA4GH Beacon 协议中存在的隐私风险。现有 Beacon 协议允许研究人员查询某个基因组变异是否在参与队列中被观察到，并