Proceedings of the Third BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP

Afra Alishahi (Editor), Yonatan Belinkov (Editor), Grzegorz Chrupała (Editor), Dieuwke Hupkes (Editor), Yuval Pinter (Editor), Hassan Sajjad (Editor)

Research output: Book/ReportAnthologypeer-review


BlackboxNLP is the workshop on analyzing and interpreting neural networks for NLP.
In the last few years, neural networks have rapidly become a central component in NLP systems. The improvement in accuracy and performance brought by the introduction of neural networks has typically come at the cost of our understanding of the system: How do we assess what the representations and computations are that the network learns? The goal of this workshop is to bring together people who are attempting to peek inside the neural network black box, taking inspiration from machine learning, psychology, linguistics, and neuroscience.
In this third edition of the workshop, hosted by the 2020 conference on Empirical Methods in Natural Language Processing (EMNLP), we accepted 31 archival papers and 9 extended abstracts. The workshop also provided a platform for authors of EMNLP-Findings papers to present their work as a poster at the workshop. Lastly, for the first time, BlackboxNLP included a shared interpretation mission. One paper submitted to this mission has a demo presentation, of the interpretability library diagnnose, and is included as the last paper in these proceedings (submission number 70).
BlackboxNLP would not have been possible without the dedication of its program committee. We would like to thank them for their invaluable effort in providing timely and high-quality reviews on a short notice. We are also grateful to our invited speakers for contributing to our program
Original languageEnglish GB
Place of PublicationOnline
PublisherAssociation for Computational Linguistics
StatePublished - 1 Nov 2020


Dive into the research topics of 'Proceedings of the Third BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP'. Together they form a unique fingerprint.

Cite this