Search papers, labs, and topics across Lattice.
3
0
6
Reward models can be made more robust to spurious cues like length and sycophancy by explicitly training them to understand the *intent* behind a prompt.
Achieve formally certified collision risk guarantees for robot manipulators in complex, uncertain environments with a novel risk-bounded motion planning framework.
Decoupling nuclei detection and classification in histopathology unlocks significant performance gains in pathology foundation models while drastically reducing the number of trainable parameters.
