2 papers
cs.DC2026
Tile-Level Activation Overlap for Efficient LLM Inference
Abhinav Jangda, Tyler Sorensen, Sebastian Burckhardt +3
SwiGLU is the dominant MLP activation in modern large language models, yet its intermediate tensor materialization costs 9-37% of MLP execution time. We present two complementary C…
physics.comp-ph2023
KDE-Based Coarse-graining of Semicrystalline Systems with Correlated Three-body Intramolecular Interaction
Jianlan Ye, Vipin Agrawal, Minghao Liu +2
We present an extension to the iterative Boltzmann inversion method to generate coarse-grained models with three-body intramolecular potentials that can reproduce correlations in s…