6 papers
GRACE: Generation-aware latent compression for efficient video generation
Jiyoung Kim, Paul Hyunbin Cho, Jisu Nam +5
Highly compressed video autoencoders offer an effective way to accelerate video diffusion models, as the Diffusion Transformer (DiT) operates on far fewer tokens. However, such aut…
Differential Refresh of Materialized Derived State: From a Single-Age Equivalence Limit to an Optimal Per-Partition Allocation, with a Real-System Validation on Materialized Views
Amit Rajula
Materialized views, cached aggregates, and trained models are all derived state: query results precomputed from base relations that begin to age the instant they are built, because…
Adaptive Randomized Neural Networks with Locally Activation Function: Theory and Algorithm for Solving PDEs
Ran Bi, Weibing Deng
This paper establishes an approximation theory and develop an adaptive computational framework for randomized neural networks (RaNNs). For RaNNs of the form $\sum_{i=1}^{N} W_i σ(A…
A Two-stage Adaptive Lifting PINN Framework for Solving Viscous Approximations to Hyperbolic Conservation Laws
Yameng Zhu, Weibing Deng, Ran Bi
Training physics-informed neural networks (PINNs) for hyperbolic conservation laws near the inviscid limit is challenging because shock discontinuities invalidate classical strong-…
Extended Interface Physics-Informed Neural Networks Method for Moving Interface Problems
Ran Bi, Weibing Deng, Yameng Zhu
Physics-informed neural networks (PINNs) have emerged as an effective class of mesh-free methods for solving partial differential equations (PDEs), particularly on complex geometri…
XI-DeepONet: An operator learning method for elliptic interface problems
Ran Bi, Jingrun Chen, Weibing Deng
Scientific computing has been an indispensable tool in applied sciences and engineering, where traditional numerical methods are often employed due to their superior accuracy guara…