Track new results across pure and applied mathematics.
Search papers, labs, and topics across Lattice.
Affiliation:
2
0
6
8
A strong LLM can review grounded drafts from a weak model, retaining most task performance while halving latency and memory traffic.
Dynamically recomputing embeddings at each denoising step in diffusion transformers halves training compute while achieving state-of-the-art FID scores.