Search papers, labs, and topics across Lattice.
2
0
6
0
High-fidelity image synthesis does not require paired text from day one: pre-training visual priors on uncaptioned images before multimodal alignment beats conventional joint training pipelines to establish a new open-source DiT benchmark.
RODS synthesizes new training data on-the-fly, enabling agents to maintain high performance with 20x fewer trajectories than traditional methods.
