Search papers, labs, and topics across Lattice.
Affiliation:
2
0
3
Reinforcement learning with sparse autoencoder features (RL-SAE) unlocks precise, interpretable, and composable design of intrinsically disordered protein regions, activating 90% of targeted functional features compared to 24% with activation steering.
Mean force matching slashes the data requirements for training coarse-grained molecular dynamics potentials by 50x while boosting accuracy on unseen proteins.
