Search papers, labs, and topics across Lattice.
3
0
5
Post-training strategies can reshape biological reasoning models in unexpected ways, revealing that more supervision doesn't always mean better generalization.
Decentralized AI agents, left to their own devices, can outperform centralized approaches in scientific discovery, finding better solutions faster across diverse domains like biomedicine and language model training.
Generalize transport maps to unseen source and target distributions by conditioning on learned distribution embeddings, unlocking applications in single-cell genomics, mass cytometry, and more.
