Call For Papers
First Call for Papers
Second Workshop on Ethics in Natural Language Processing
NAACL 2018, New Orleans June 5 or 6 2018

     NLP is a rapidly maturing field. NLP technologies now play a role in business applications and decision processes that affect billions of people on a daily basis.

However, increasing amounts of data and computational power also mean increased responsibility and new questions for researchers and practitioners.

     This one-day, interdisciplinary workshop will bring together researchers and practitioners in NLP with researchers in the humanities, social sciences, public policy, and law to identify and discuss some of the most pressing issues surrounding ethics in NLP, for example:

     • Are we inadvertently building unfair biases into our     
     data sets and models?
     • What information is ethical to infer from user data?
     • How can we prioritize accountability and transparency
     • What are the big-picture ethical consequences and
     implications of our work?

The workshop consists of:

       • invited talks (given by researchers in NLP but also in
       AI, philosophy, the social sciences, or law)
       • contributed talks and posters
       • panel discussions with NLP researchers, ethicists,

       lawyers, and industry practitioners.

We invite submissions on any area of NLP that touches on the following topics:
       • Bias in NLP models (e.g., reporting bias, implicit bias).
       • Exclusion and inclusion (e.g., exclusion of certain

       groups or beliefs, how/when to include stakeholders
       and representatives for the user population to be
       • Overgeneralization (e.g., making false classifications

       on tasks including authorship attribution, NER,
       knowledge base population).
       • Exposure (e.g., underrepresentation/
       overrepresentation of languages or groups).
       • Dual use (e.g., the positive and negative aspects of

       NLP applications, the close relationship between
       government and industry interests and NLP research).
       • Privacy protection (e.g., anonymization of biomedical        documents, best practices for researchers in industry          to ensure the privacy of their users’ data, educating the        public about how much industry and government may        know about them, privacy protection for data

       annotated with non-linguistic features such as
       • Any other topic which concerns ethical considerations        in NLP.

Organizing Committee:
         Dirk Hovy, Bocconi University
Margaret Mitchell, Google Research and Machine Intelligence
       Shannon Spruit, Technical University Delft
       Michael Strube, Heidelberg Institute for Theoretical Studies