16 citations · 36 across the 9 of their papers we have counts for
9 papers
From Objects to Events: Unlocking Complex Visual Understanding in Object Detectors via LLM-guided Symbolic Reasoning
Yuhui Zeng, Haoxiang Wu, Wenjie Nie +6
Current object detectors excel at entity localization and classification, yet exhibit inherent limitations in event recognition capabilities. This deficiency arises from their arch…
Decoupled Rationalization with Asymmetric Learning Rates: A Flexible Lipschitz Restraint
Wei Liu, Jun Wang, Haozhao Wang +5
A self-explaining rationalization model is generally constructed by a cooperative game where a generator selects the most human-intelligible pieces from the input text as rationale…
CSGCL: Community-Strength-Enhanced Graph Contrastive Learning
Han Chen, Ziwen Zhao, Yuhua Li +3
Graph Contrastive Learning (GCL) is an effective way to learn generalized graph representations in a self-supervised manner, and has grown rapidly in recent years. However, the und…
Margin-Based Few-Shot Class-Incremental Learning with Class-Level Overfitting Mitigation
Yixiong Zou, Shanghang Zhang, Yuhua Li +1
Few-shot class-incremental learning (FSCIL) is designed to incrementally recognize novel classes with only few training samples after the (pre-)training on base classes with suffic…
Annotation Efficient Person Re-Identification with Diverse Cluster-Based Pair Selection
Lantian Xue, Yixiong Zou, Peixi Peng +2
Person Re-identification (Re-ID) has attracted great attention due to its promising real-world applications. However, in practice, it is always costly to annotate the training data…
Annotation-Efficient Untrimmed Video Action Recognition
Yixiong Zou, Shanghang Zhang, Guangyao Chen +3
Deep learning has achieved great success in recognizing video actions, but the collection and annotation of training data are still quite laborious, which mainly lies in two aspect…
Revisiting Mid-Level Patterns for Cross-Domain Few-Shot Recognition
Yixiong Zou, Shanghang Zhang, JianPeng Yu +2
Existing few-shot learning (FSL) methods usually assume base classes and novel classes are from the same domain (in-domain setting). However, in practice, it may be infeasible to c…
Compositional Few-Shot Recognition with Primitive Discovery and Enhancing
Yixiong Zou, Shanghang Zhang, Ke Chen +3
Few-shot learning (FSL) aims at recognizing novel classes given only few training samples, which still remains a great challenge for deep learning. However, humans can easily recog…
ODN: Opening the Deep Network for Open-set Action Recognition
Yu Shu, Yemin Shi, Yaowei Wang +3
In recent years, the performance of action recognition has been significantly improved with the help of deep neural networks. Most of the existing action recognition works hold the…