Search papers, labs, and topics across Lattice.
3
0
7
Text world models can transform LLM-based agents from reactive responders to proactive planners, fundamentally changing how they interact with complex environments.
LVLMs can now perform visual search far more effectively thanks to a clever decoding strategy that harmonizes pre- and post-training capabilities.
Video-LLMs can now stream more effectively: WeaveTime teaches them to perceive temporal order and focus dynamically on relevant history, boosting accuracy and cutting latency without requiring architectural changes.
