Track new results across pure and applied mathematics.
Search papers, labs, and topics across Lattice.
Affiliation:
4
0
5
5
Rank actionable recommendation model hypotheses with 79-88% compute savings by predicting full-scale performance from small-scale proxies.
PLES can uncover optimal hyperparameter scaling laws for LLMs using a fraction of the resources typically required, revolutionizing hyperparameter tuning efficiency.
Forget exotic attention mechanisms – MobileLLM-Flash achieves up to 1.8x faster LLM prefill on mobile CPUs by smartly pruning and adapting existing architectures for on-device use.
Forget handcrafted kernels: Empirical GPs learn flexible, data-driven priors directly from historical data, unlocking richer covariance structures.