1. Editing Knowledge in Text-to-Image Generation Models
      NVIDIA AI Research; Berkeley University
    1. On Localization in Language Models
      Google Research India; IBM Research Tel Aviv; Google Research Tel Aviv; TU Darmstadt; Düsseldorf University; Bar Ilan University; Simons Institute; Microsoft Research

  • February, 2021. Interpretability and Robustness in Natural Language Processing, AAAI New Faculty Highlights

  • August, 2020. Interpretability and Other Highlights from NLP, Workshop on Decoding Communication in Nonhuman Species
    Simons Institute, UC Berkeley

  • Winter-Spring, 2020. Causal Mediation Analysis for Interpreting Neural NLP: The Case of Gender Bias
    Stanford University; UC Berkeley; UMass Amherst; Salesforce; Google; Amazon; NYU (Video)

  • January-February, 2019. Deep Learning Models for Language: What they learn, where they fail, and how to make them more robust
    Hebrew University, Jerusalem
    Technion, Haifa
    Weizmann Institute, Rehovot
    Carnegie Mellon University, Pittsburgh
    University of Pennsilvenia, Philadelphia

  • October, 2018. Internal Representations in Neural Machine Translation
    Amazon MT team, Pittsburgh

  • February-March, 2018. Internal Representations in Deep Learning for Language and Speech Processing
    CLSP, Johns Hopkins University, Baltimore
    University of Washington, Seattle
    Allen Institute for Artificial Intelligence, Seattle (Video)
    Toyota Technological Institute at Chicago, Chicago
    Radcliffe Institute for Advanced Study, Harvard, Cambridge, MA

  • November, 2017. Understanding Internal Representations in Deep Learning Models for Language and Speech Processing
    Machine Learning for Language, NYU, New York

  • June, 2017. On Learning Form and Meaning in Neural Machine Translation Models
    Computational Data Science Seminar, Technion, Haifa

  • May, 2017. What do Neural Machine Translation Models Learn about Morphology?
    Data Science Summit Europe, Jerusalem

  • May, 2017. On Learning Form and Meaning in Neural Machine Translation Models
    CompLang Discussion Group, MIT, Cambridge, MA

  • January, 2017. Language Technologies for Arabic: Historical Documents, Web Forums, and Machine Translation
    Qatar Computing Research Institute, Doha

  • June, 2016. A Computational Analysis of Judeo-Arabic Translations of the Passover Hagaddah
    International Jewish Languages Conference, Hebrew University of Jerusalem, Jerusalem

  • August, 2015. Deep Learning for Sentence Representation
    IBM Research, Tel Aviv

  • January, 2015. Exploring Compositional Architectures and Word Vector Representations for Prepositional Phrase Attachment
    Tel Aviv University, Tel Aviv

  • October, 2013. arTenTen: A Web Corpus for Arabic Varieties
    AIMA, Emory University, Atlanta, GA

  • February, 2012. A Dialectometrical Analysis of Judeo-Arabic Translations of the Passover Haggadah
    Jil Jadid, University of Texas at Austin, Austin, TX

  • February, 2012. The Palestinian Dialect of Jisir izZarga
    Emory University, Atlanta, GA