George Mason is a University Site for the NACLO 2026 Open Round. Please see below for details and instructions to participants.
George Mason is a University Site for the NACLO 2025 Open Round. Please see below for details and instructions to participants.
The advent of digital era has made natural language processing (NLP) technology both promising and important for disaster response. Language technologies can, for instance, be used for triaging the need for immediate assistance based on text messages, translating vast amounts of data related to ongoing pandemics, or engage in conversations to guide natural disaster victims. However, these advances are only limited to a few dozen of the more than 6500 languages spoken or signed in the world today, neglecting millions of people and widening the "digital divide", effectively reducing the technologies' real-world utility.
This is a post regarding our paper that will be presented at ACL 2020. tl;dr: You can use previously published results to get an estimation of the performance on a new experiment, before running it! We describe NLP experiments with a set of interpretable features, and train prediction models which are actually pretty good at producing such estimations, and show how they can be useful in various scenarios.
This is a post regarding our paper that got accepted at ACL 2020.
Word embeddings are ubiquitous in modern NLP, from static ones (like word2vec or fasttext) to contextual representations obtained from ELMo, BERT, and other models. One very interesting property of learned word embeddings is that, since they (ideally) model the syntactic or semantic relations between words, the embedding space that we learn in two different languages will have somewhat similar properties and structure.
tl;dr: Be careful when reporting averages for multilingual benchmarks, especially if making claims about multilinguality. In addition, averaging by language family can provide additional insights.