Search papers, labs, and topics across Lattice.
University of Chicago
4
0
10
INFUSER outperforms a frozen 32B model with just an 8B co-evolving generator, showcasing the power of adaptive question generation in self-evolution.
Achieve verifiable clinical interpretation by grounding radiology reports to 3D CT volumes with a novel graph-guided lesion grounding framework that outperforms existing multimodal foundation models.
Layer-selective rehearsal and rapid recovery strategies can boost model performance in federated learning by over 30% in real-world applications.
Stop optimizing generative engines in isolation: MAGEO learns reusable editing strategies that dramatically improve visibility and citation fidelity across diverse engines.
