Track the frontier, from condensed matter to cosmology.
Search papers, labs, and topics across Lattice.
Affiliation:
3
0
2
Slash recommendation model exploration costs by 80% with a framework that accurately predicts full-scale performance from small-scale proxies.
LLMs can learn from instructions and tasks far beyond their pretraining support by leveraging demonstration evidence to overcome instruction shifts and decomposing complex goals into familiar subtasks.
Few-step diffusion models can achieve state-of-the-art 4-NFE generation with simple on-policy velocity regression and a novel residual proxy, matching multi-step teacher performance without complex auxiliary models.