Large language models (LLMs) have shown remarkable capabilities in language
understanding and generation. However, such impressive capability typically comes with a …

Structural pruning enables model acceleration by removing structurally-grouped parameters
from neural networks. However, the parameter-grouping patterns vary widely across …

Diffusion models are powerful, but they require a lot of time and data to train. We propose
Patch Diffusion, a generic patch-wise training framework, to significantly reduce the training …

Despite the visually-pleasing results achieved, the massive computational cost has been a
long-standing flaw for diffusion probabilistic models (DPMs), which, in turn, greatly limits …

Abstract Diffusion Probabilistic Models (DPMs) have recently demonstrated impressive
results on various generative tasks. Despite its promises, the learned representations of pre …

Adapter-style efficient transfer learning (ETL) has shown excellent performance in the tuning
of vision-language models (VLMs) under the low-data regime, where only a few additional …

Vision Transformer has demonstrated impressive success across various vision tasks.
However, its heavy computation cost, which grows quadratically with respect to the token …

Large-scale vision-language models (VLMs) pre-trained on billion-level data have learned
general visual representations and broad visual concepts. In principle, the well-learned …

The ultimate goal of Dataset Distillation is to synthesize a small synthetic dataset such that a
model trained on this synthetic set will perform equally well as a model trained on the full …

Video grounding, the process of identifying a specific moment in an untrimmed video based
on a natural language query, has become a popular topic in video understanding. However …