University of California, Riverside

Antibody CDR design model fusing diffusion with a GFlowNet objective so binding energy is optimized during training rather than by post-hoc RL.

Constrained RNA sequence design that turns an encoder-only RNA language model into a generator by greedy gradient search over nucleotides.

RNA language model that switches between nucleotide and byte-pair tokenization by input length, so one 117M encoder handles sequences of any length.

Fine-tuned DNA language model that predicts differentially methylated cytosines in TET and DNMT3 knockout stem cells from sequence alone.