Learning discriminative image representations plays a vital role in long-tailed image
classification because it can ease the classifier learning in imbalanced cases. Given the …

Semi-supervised learning on class-imbalanced data, although a realistic problem, has been
under studied. While existing semi-supervised learning (SSL) methods are known to perform …

The problem of class imbalanced data is that the generalization performance of the classifier
deteriorates due to the lack of data from minority classes. In this paper, we propose a novel …

The heavy reliance on data is one of the major reasons that currently limit the development
of deep learning. Data quality directly dominates the effect of deep learning models, and the …

The current evaluation protocol of long-tailed visual recognition trains the classification
model on the long-tailed source label distribution and evaluates its performance on the …

Existing visual question answering methods often suffer from cross-modal spurious
correlations and oversimplified event-level reasoning processes that fail to capture event …

We present a novel counterfactual framework for both Zero-Shot Learning (ZSL) and Open-
Set Recognition (OSR), whose common challenge is generalizing to the unseen-classes by …

Self-supervised learning (SSL) is a scalable way to learn general visual representations
since it learns without labels. However, large-scale unlabeled datasets in the wild often have …

We uncover an ever-overlooked deficiency in the prevailing Few-Shot Learning (FSL)
methods: the pre-trained knowledge is indeed a confounder that limits the performance. This …

We propose a causal framework to explain the catastrophic forgetting in Class-Incremental
Learning (CIL) and then derive a novel distillation method that is orthogonal to the existing …