8 papers · 1 filter
Poisoning the corpora of protein design systems during pre-training and retrieval
Charalampos Koilakos, Kimon-Antonios Provatas, Michail Patsakis +3
Protein foundation models increasingly support the generation of novel biological sequences, yet their reliance on large public datasets creates opportunities for adversarial manip…
Tracing model-generated DNA with position-independent watermarking
Kimon Antonios Provatas1, Christos Galanopoulos, Ilias Georgakopoulos-Soares
Genomic language models can write synthetic DNA that carries no record of its origin. A generation-time watermark could provide a provenance signal, if a verifier can detect it usi…
Auditing System-1 Models on Biosecurity-Relevant Benchmarks: Calibration, Selective Prediction, and Permutation Instability in a Non-Generative Model
Kimon Antonios Provatas, Ilias Georgakopoulos-Soares
Non-generative "System-1" models return structured probabilistic decisions in a single forward pass, without autoregressive decoding, at a small fraction of the inference cost of a…
Feasibility of Homomorphic Inference for a Genomic Foundation Model
Christos Galanopoulos, Kimon Antonios Provatas, Ilias Georgakopoulos-Soares
Human genomic sequences can identify individuals, cannot be replaced after disclosure, and are the inputs that genomic foundation models are designed to interpret. We assess whethe…
bioETH-Beacon: A Confidential On-Chain Genomic Beacon with Encrypted Counts, Filters, and Bounded Noise over a Fully Homomorphic EVM
Christos Galanopoulos, Kimon Antonios Provatas, Ilias Georgakopoulos-Soares
The Global Alliance for Genomics and Health (GA4GH) Beacon protocol lets researchers ask whether a genomic variant has been observed in a participating cohort and receive aggregate…
bioETH-PRS: Confidential Polygenic Risk Scoring without a Trusted Evaluator via Fully Homomorphic Encryption on a Programmable Blockchain
Kimon Antonios Provatas, Christos Galanopoulos, Ilias Georgakopoulos-Soares
Polygenic risk scores (PRSs) aggregate genetic effect estimates to predict disease susceptibility, yet clinical deployment often exposes raw genotype data to third-party compute in…
BioVeil MATRIX: Uncovering and categorizing vulnerabilities of agentic biological AI scientists
Kimon Antonios Provatas, Avery Self, Ioannis Mouratidis +1
Agentic AI scientists equipped with domain-specific tools are rapidly entering scientific workflows across disciplines, with especially strong uptake in the life sciences where the…
BEDCrypt: Privacy-preserving interval analytics with homomorphic encryption
Kimon Antonios Provatas, Ilias Georgakopoulos-Soares
Motivation. Genomic data and derived interval datasets can carry sensitive information, and the analysis itself can reveal an analyst's intent. As genomic workloads are increasingl…