Search papers, labs, and topics across Lattice.
5
0
11
3
LLM distillation can achieve 18.55% higher math accuracy and 13.8 points better ALFWorld success by intelligently branching student trajectories at uncertain decision points, not just at turn starts.
A 260M-parameter looped diffusion model achieves performance surpassing a 6.5x larger non-looped counterpart with significantly less inference compute.
OpenMobile proves that high-performing mobile agents can be trained on entirely synthetic, open-source data, closing the gap with closed-source models and enabling broader research.
Generate safety-critical driving scenarios with full trajectory control, even *beyond* your training data, using RL to fine-tune a video diffusion model.
Turning past programming failures into reusable knowledge boosts automated repair performance by 3.7% on a multimodal benchmark.
