Search papers, labs, and topics across Lattice.
3
0
6
Heterogeneous LLMs can fix each other's dead-end zero-reward rollouts in RLVR through off-policy trajectory swapping, boosting math reasoning by up to 4.5 points without extra compute or a stronger teacher model.
Ditching autoregressive token generation in favor of continuous flow matching slashes tool-use latency by over 50脳 while beating proprietary MLLMs at image retouching.
This work introduces SpatialBlock-15k, a synthetic dataset of 15,000 block-stacking problems covering 3D-to-2D projection, viewpoint transformation, and structural combination and proposes a novel paradigm inspired by human cognitive development: learning foundational spatial skills through structured block-manipulation tasks.
