DPLM-Evo introduces an evolutionary discrete diffusion framework with explicit edit prediction and contextual noising that claims SOTA single-sequence mutation effect prediction on ProteinGym while supporting variable-length evolution simulation.
Simulating 500 million years of evolution with a language model
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
No citing papers match the current filters.