Series
Challenges and Opportunities in Evaluating Progress in NLP
Abstract: The past few years have seen remarkable advances in NLP, as evidenced both by continued and rapid gains on benchmark tasks, as well as by the increasing prominence of real NLP systems in the wild. In assessing such progress, however, it is important to ask not only what system achieves the best performance, but how it achieves that level of performance, how much we can trust the evaluation, and what the consequences of deploying such a system might be.
A Typology of Ethical Risks in Language Technology with an Eye Towards Where Transparent Documentation Can Help
Bilingualism is the norm, not the exception: How using two or more languages shapes the mind and the brain and reflects society itself
Join us for a panel discussion with researchers who transitioned their lab-based research online over the summer.
Arynn Byrd (HESP) is using a web app to remotely administer a sentence-picture matching task with children. The goal of the study is to evaluate if there are differences in how children who speak different dialects use verb knowledge to disambiguate the sentence. Her web app displays pictures, plays audio, and gathers reaction times and accuracy using the child's tablet.
This semester we're taking a break from the usual Language Science Lunch Talks to reimagine what we can do with this weekly opportunity for interdisciplinary discussion. We want to create an environment where all members of our community feel comfortable expressing ideas, learning to communicate with people from different discpiplines, and maintaining relationships with peers from other departments. We recognize that the traditional academic talk format may not be the best way to achieve these goals.
Structuring Posterior Inference: Unsupervised methods to find gendered words and generate sentiment lexica
Pages
