Search papers, labs, and topics across Lattice.
Google DeepMind, Harvard University, Google Research
1
0
3
Post-training strategies can reshape biological reasoning models in unexpected ways, revealing that more supervision doesn't always mean better generalization.
