Track the frontier, from condensed matter to cosmology.
Search papers, labs, and topics across Lattice.
4
0
8
18
Visual generation supervision injects significant, yet often inaccessible, knowledge into UMMs, requiring novel alignment strategies to unlock its full potential for text-based reasoning.
Multimodal perception is no longer just an add-on: GLM-5V-Turbo bakes it directly into the core of reasoning, planning, and action.
Today's best visual coding agents still struggle to build complete websites from scratch, as revealed by a new benchmark spanning UI generation to full-stack development.
A compact 0.9B multimodal model, GLM-OCR, achieves state-of-the-art document understanding by predicting multiple tokens at once, boosting decoding throughput without blowing up memory.