Analyzing and interpreting neural networks for NLP: A report on the first BlackboxNLP workshop

Afra Alishahi, Grzegorz Chrupała, Tal Linzen

    Research output: Contribution to journalArticlepeer-review


    The Empirical Methods in Natural Language Processing (EMNLP) 2018 workshop BlackboxNLP was dedicated to resources and techniques specifically developed for analyzing and understanding the inner-workings and representations acquired by neural models of language. Approaches included: systematic manipulation of input to neural networks and investigating the impact on their performance, testing whether interpretable knowledge can be decoded from intermediate representations acquired by neural networks, proposing modifications to neural network architectures to make their knowledge state or generated output more explainable, and examining the performance of networks on simplified or formal languages. Here we review a number of representative studies in each category.

    Original languageEnglish (US)
    Pages (from-to)543-557
    Number of pages15
    JournalNatural Language Engineering
    Issue number4
    StatePublished - Jul 1 2019


    • interpretability
    • natural language processing
    • neural networks

    ASJC Scopus subject areas

    • Software
    • Language and Linguistics
    • Linguistics and Language
    • Artificial Intelligence

    Fingerprint Dive into the research topics of 'Analyzing and interpreting neural networks for NLP: A report on the first BlackboxNLP workshop'. Together they form a unique fingerprint.

    Cite this