Search papers, labs, and topics across Lattice.
3
0
5
3
GPU kernel optimization agents can regress by up to 19% when transferring distilled knowledge across DSLs, revealing that source attempt quality, not just success, dictates transfer efficacy.
LLMs can now generate mobile kernels that outperform native libraries 27.4% of the time, thanks to a new multi-agent system that tackles compilation failures and hallucinations.
Diffusion LMs can now beat autoregressive models at CUDA kernel generation, thanks to a new dataset and reinforcement learning approach.
