DIORA: An unsupervised tree parser and LM
This is my primary area of work with Prof. Andrew McCallum and Prof. Mohit Iyyer. We are working on improving DIORA (Deep Inside Outside Representations using Autoencoders), a fully unsupervised method for inducing tree-structure in english sentences based on dynamic programming. I’m currently working on approximating the dynamic program using representating learning and knowledge distillation.
In the past, I worked on a variant of DIORA called S-DIORA, where the soft dynamic program is replaced by a more exact, hard version. This work was in collaboration with Andrew Drozdov and it was published at EMNLP 2020.