For high-resource languages like English, text classification is a
well-...
Training deep neural networks (DNNs) with weak supervision has been a ho...
Learning semantically meaningful sentence embeddings is an open problem ...
Incorrect labels in training data occur when human annotators make mista...
Welcome to WeaSuL 2021, the First Workshop on Weakly Supervised Learning...
Distant supervision allows obtaining labeled training corpora for
low-re...
Distant and weak supervision allow to obtain large amounts of labeled
tr...
Current developments in natural language processing offer challenges and...
Multilingual transformer models like mBERT and XLM-RoBERTa have obtained...
Fine-tuning pre-trained contextualized embedding models has become an
in...
We study and quantify the generalization patterns of multitask learning ...
The lack of labeled training data has limited the development of natural...
In low-resource settings, the performance of supervised labeling models ...
Popular word embedding methods such as word2vec and GloVe assign a singl...
In this paper, we address the problem of effectively self-training neura...
Manually labeled corpora are expensive to create and often not available...