Ana Marasovic on “NLP’s generalization problem, and how researchers are tackling it”

Ana Marasović is a writer for the online magazine The Gradient and a final year PhD student in AIPHES.

In her article she breaks down the recently discussed "NLP generalization problem" and shares approaches and ideas of Yejin Choi, Dan Roth, and Percy Liang outlined at the NAACL workshop on generalization in Deep Learning and NLP.

In her own words, this discussion is a healthy sign that AI researchers are very much clear-eyed about the limitations of Deep Learning, and working to address them.