Search papers, labs, and topics across Lattice.
Affiliation:
3
0
2
Streaming video models consistently fail at continuous counting because they cannot distinguish re-observed entities from new instances鈥攁 failure mode resolved by framing online counting as differentiable state-machine updates over persistent visual trajectories.
Tossing out entire visual plans when robot execution drifts is needlessly wasteful: selectively branching from intermediate generative states preserves valid task structure and drives closed-loop manipulation success up to 84.8%.
Generation-guided training can significantly enhance multimodal understanding in MLLMs without any inference overhead.
