Search papers, labs, and topics across Lattice.
Affiliation:
3
0
7
17
Spatial coding agents can be distilled into standalone MLLMs, achieving superior spatial reasoning and generalization without external tools.
WnW reduces GPU memory usage to 20% of audio tokens without sacrificing accuracy, challenging the limitations of existing KV cache methods in long-form speech processing.
OvisOCR2 outperforms traditional pipeline methods, achieving a groundbreaking score of 96.58 in end-to-end document parsing.
