Welcome I am a postdoctoral researcher at the Paul G. Allen School of Computer Science & Engineering department of the University of Washington working with Professor Noah Smith. My current research on unified, structure-aware and sample efficient models of natural language is funded by SNSF. Previously, I did a postdoc with Doctor James Henderson at the Idiap Research Institute's natural language understanding group and earned my doctoral degree at EPFL with Professor Andrei Popescu-Belis and Professor Hervé Bourlard. I have also interned at Columbia University and received my master's and engineer's degrees from the University of the Aegean.

News  Show more

• Jun 2020 — Paper on Fast Autoregressive Transformers with Linear Attention to appear at ICML 2020.
• Jun 2020 — Preprints on speeding-up transformers with Linear Attention and a Deep-Shallow Model are now available.
• Sep 2019 — Joined Noah's ARK group and the NLP group at the University of Washington.
• May 2019 — Papers on Generalized Zero-Shot Learning and Sense-Aware NMT will be presented at ACL 2019.
• Apr 2019 — Paper on Deep Residual Output Layers for Neural Language Generation was accepted at ICML 2019.
• Jan 2019 — Paper on A Generalized Input-Label Embedding for Text Classification to appear in TACL.
• Dec 2018 — Won a two-year postdoc fellowship by SNSF to work with prof. Noah Smith at the University of Washington!
Oct 2018 — Paper on Integrating Weakly Supervised Word Sense Disambiguation into Neural MT to appear in TACL.
Aug 2018 — Paper on Document-Level Neural MT with Hierarchical Attention Networks was accepted at EMNLP 2018.
Aug 2018 — Paper on Joint Input-Output Embeddings for Neural MT was accepted at WMT 2018.
Jun 2018 — Gave an invited talk on Basics and (Modern) Methods for NLP at Digital School of UnternehmerTUM.
Feb 2018 — Paper on Self-attentive Residual Decoder for Neural MT was accepted at NAACL 2018.
Nov 2017 — Won the best paper award at IJCNLP 2017!
Sep 2017 — Paper on Multilingual Hierarchical Attention Networks was accepted at IJCNLP 2017.
Jun 2017 — Paper on Sense-aware Statistical Machine Translation was accepted at WMT 2017.
Feb 2017 — Journal paper on Explicit Document Modeling through Multiple-Instance Learning was accepted at JAIR.
Dec 2016 — Gave a lecture on Multilingual Word Sequence Learning for the HLT course at EPFL.
Nov 2016 — Attended EMNLP 2016 conference in Austin, Texas.
Oct 2016 — Gave a lecture on Word Representation Learning for the HLT course at EPFL.

Present

2019

2016

2014

2011

Contact me at: npappas@cs.washington.edu
Scholar / Github / Twitter