PPL
Weili Guan
Papers (5)
DASH: Fast Differentiable Architecture Search for Hybrid Attention in Minutes on a Single GPUVISTA: Technical Report for the Ego4D Short-Term Object Interaction Anticipation at EgoVis 2026PulseCol: Periodically Refreshed Column-Sparse Attention for Accelerating Diffusion Language ModelsMARS: Technical Report for the CASTLE Challenge at EgoVis 2026Detecting Deepfakes via Hamiltonian Dynamics