u-LLaVA: Unifying Multi-Modal Tasks via Large Language Model

1 code implementation • 9 Nov 2023 • Jinjin Xu, Liwu Xu, Yuzhe Yang, Xiang Li, Fanyi Wang, Yanchun Xie, Yi-Jie Huang, Yaqian Li

Recent advancements in multi-modal large language models (MLLMs) have led to substantial improvements in visual understanding, primarily driven by sophisticated modality alignment strategies.

Instruction Following Language Modelling +1

CLIP Brings Better Features to Visual Aesthetics Learners

no code implementations • 28 Jul 2023 • Liwu Xu, Jinjin Xu, Yuzhe Yang, YiJie Huang, Yanchun Xie, Yaqian Li

Specifically, we first integrate and leverage a multi-source unlabeled dataset to align rich features between a given visual encoder and an off-the-shelf CLIP image encoder via feature alignment loss.

Mixed Sample Augmentation for Online Distillation

no code implementations • 24 Jun 2022 • Yiqing Shen, Liwu Xu, Yuzhe Yang, Yaqian Li, Yandong Guo

Mixed Sample Regularization (MSR), such as MixUp or CutMix, is a powerful data augmentation strategy to generalize convolutional neural networks.

Data Augmentation Knowledge Distillation

Personalized Image Aesthetics Assessment with Rich Attributes

no code implementations • CVPR 2022 • Yuzhe Yang, Liwu Xu, Leida Li, Nan Qie, Yaqian Li, Peng Zhang, Yandong Guo

To solve the dilemma, we conduct so far, the most comprehensive subjective study of personalized image aesthetics and introduce a new Personalized image Aesthetics database with Rich Attributes (PARA), which consists of 31, 220 images with annotations by 438 subjects.

Self-Distillation from the Last Mini-Batch for Consistency Regularization

1 code implementation • CVPR 2022 • Yiqing Shen, Liwu Xu, Yuzhe Yang, Yaqian Li, Yandong Guo

Afterwards, the former half mini-batch distills on-the-fly soft targets generated in the previous iteration.

Knowledge Distillation