Open access
Sep 2026
Generative Language Modeling for Antibody CDR Grafting and Alignment-driven De Novo Design
GenCDR, a family of LLaMa-based autoregressive language models that read all frameworks as a conditioning prompt and generate all CDRs jointly as a variable-length response, makes CDR likelihoods a clean, separable target for reward attribution, achieves the highest CDR recovery among autoregressive models.
Ferran Gonzalez Hernandez, Oliver M. Turnbull, Maliha Sultana et al.
· bioRxiv · 0 citations