Elias Stengel-Eskin

Elias Stengel-Eskin

PhD Student

Johns Hopkins University

Biography

I am a fourth-year Ph.D. student at the Center for Language and Speech Processing supervised by Benjamin Van Durme, where I am generously supported by an NSF Graduate Research Fellowship. My work focuses mostly on transforming text into representations of its meaning. Beyond building models to do this transformation, I also spend time thinking/writing about meaning, especially as it relates to NLP. Most recently, I have been working on multimodal grounding and human-robot interaction, as well as semantic parsing for Universal Decompositional Semantics. Before starting my Ph.D., I received my B.A.&Sc. with First Class Honours in Cognitive Science from McGill University, focusing in computer science and linguistics. While at McGill, I worked as a research assistant at the Montreal Language Modeling Lab (MLML), now MCQLL. I wrote my honours thesis (supervised by Timothy O’Donnell) on a variational inference algorithm for a model of language acquisition.

Interests

  • Semantic Parsing
  • Multimodal Grounding
  • Compositionality
  • Vagueness

Education

  • PhD in Computer Science, 2018-present

    Johns Hopkins University

  • MSE in Computer Science, 2018-2021

    Johns Hopkins University

  • BA&Sc in Cognitive Science (Honours), 2014-2018

    McGill University

Recent Publications

Quickly discover relevant content by filtering publications.

Guiding Multi-Step Rearrangement Tasks with Natural Language Instructions

Enabling human operators to interact with robotic agents using natural language would allow non-experts to intuitively instruct these …

Iterative Paraphrastic Augmentation with Discriminative Span Alignment

Joint Universal Syntactic and Semantic Parsing

While numerous attempts have been made to jointly parse syntax and semantics, high performance in one domain typically comes at the …

Human-Model Divergence in the Handling of Vagueness

While aggregate performance metrics can generate valuable insights at a large scale, their dominance means more complex and nuanced …

Universal Decompositional Semantic Parsing

We introduce a transductive model for parsing into Universal Decompositional Semantics (UDS) representations, which jointly learns to …

Recent Posts

Exploring fine-grained semantic inferences cross-lingually

Exploring fine-grained semantic inferences cross-lingually

Recent & Upcoming Presentations

Human-Model Divergence in the Handling of Vagueness

Presenting an extended abstract on some differences between people and models when handling vague predicates.

CLSP Seminar Talk: Universal Decompositional Semantic Parsing

Presented work from ACL 2020 alongside new results in the CLSP seminar

Universal Decompositional Semantic Parsing

We introduce a transductive model for parsing into Universal Decompositional Semantics (UDS) representations, which jointly learns to …

Deep Learning Fundamentals

Gave a pair of lectures on DNNs in EN.601.464 (Artificial Intelligence)

Philosophy of Language for NLP

Summarizes major questions in the philosophy of language and how they relate to various problems in NLP.

Contact

  • firstname@jhu.edu
  • DM Me