4 papers
Mitigating Backdoors via Decoy Shortcuts and Knowledge Decoupling
Zixuan Zhu, Rui Wang, Lihua Jing +1
Backdoor attacks pose a serious threat to deep neural networks, especially when training relies on third-party data, allowing adversaries to inject malicious behaviors through data…
SSTAG: Structure-Aware Self-Supervised Learning Method for Text-Attributed Graphs
Ruyue Liu, Rong Yin, Xiangzhen Bo +5
Large scale pretrained models have revolutionized Natural Language Processing (NLP) and Computer Vision (CV), showcasing remarkable cross domain generalization abilities. However,…
LDP: Generalizing to Multilingual Visual Information Extraction by Language Decoupled Pretraining
Huawen Shen, Gengluo Li, Jinwen Zhong +1
Visual Information Extraction (VIE) plays a crucial role in the comprehension of semi-structured documents, and several pre-trained models have been developed to enhance performanc…
Arbitrary Reading Order Scene Text Spotter with Local Semantics Guidance
Jiahao Lyu, Wei Wang, Dongbao Yang +2
Scene text spotting has attracted the enthusiasm of relative researchers in recent years. Most existing scene text spotters follow the detection-then-recognition paradigm, where th…