In offering this tool, we explore teachers' distinctive needs when designing chatbots to assist their teaching, and how chatbot design tools might better support them.
Characterizing these errors in easily interpretable terms gives insight into whether a classifier is prone to making systematic errors, but also gives a way to act and improve the classifier.
However, text classification in low-resource languages is still challenging due to the lack of annotated data.
Training deep neural networks (DNNs) under weak supervision has attracted increasing research attention as it can significantly reduce the annotation cost.
Learning semantically meaningful sentence embeddings is an open problem in natural language processing.
Incorrect labels in training data occur when human annotators make mistakes or when the data is generated via weak or distant supervision.
Welcome to WeaSuL 2021, the First Workshop on Weakly Supervised Learning, co-located with ICLR 2021.
Distant supervision allows obtaining labeled training corpora for low-resource settings where only limited hand-annotated data exists.
Distant and weak supervision allow to obtain large amounts of labeled training data quickly and cheaply, but these automatic annotations tend to contain a high amount of errors.
Deep neural networks and huge language models are becoming omnipresent in natural language applications.
Multilingual transformer models like mBERT and XLM-RoBERTa have obtained great improvements for many NLP tasks on a variety of languages.
Our analysis reveals that while fine-tuning indeed changes the representations of a pre-trained model and these changes are typically larger for higher layers, only in very few cases, fine-tuning has a positive effect on probing accuracy that is larger than just using the pre-trained model with a strong pooling method.
Although multitask learning has achieved improved performance in some problems, there are also tasks that lose performance when trained together.
Techniques such as distant and weak supervision can be used to create labeled data in a (semi-) automatic way.
In low-resource settings, the performance of supervised labeling models can be improved with automatically annotated or distantly supervised data, which is cheap to create but often noisy.
However, they typically cannot serve as a drop-in replacement for conventional single-sense embeddings, because the correct sense vector needs to be selected for each word.
In our experiments on Chunking and NER, this approach performs more robustly than the baselines.