Grzegorz Chrupała

News

Tweets by @gchrupala

Recent

• Open PhD position in analysis and control techniques for deep-learning spoken language applications. This vacancy is part of the InDeep consortium project. Open until filled.
• I am Area Chair for Interpretability and Analysis of Models for NLP at EMNLP 2021.
• I was a member of Madhumita Sushil's PhD defense committee. Congratulations Madhumita!
• A paper accepted to EACL 2021.
• January 2021: Teaching at the ALPS winter school on NLP.
• The consortium project InDeep: Interpreting Deep Learning Models for Text and Sound was awarded circa €2,000,000 in funding from the Dutch Research Council (NWO). I will be involved in several sub-projects, and will be leading one on analysis methods for models of spoken language. Consortium website.

People

• Bertrand Higy. Postdoc, Understanding visually grounded spoken language via multi-tasking.
• Patrick Bos. E-science engineer, Understanding visually grounded spoken language via multi-tasking.
• Christiaan Meijer. E-science engineer, Understanding visually grounded spoken language via multi-tasking.
• Chris Emmery. PhD candidate, NLP for cybersecurity.
• Lieke Gelderloos. PhD candidate, modeling language learning as a grounded and interactive process.

Alumni

• Ákos Kádár. PhD thesis: Learning Visually Grounded and Multilingual Representations

Publications

For the complete list of publications check: Google Scholar | Semantic Scholar | DBLP | ACL Anthology | ORCID

Selected papers

1. Chrupała, G. (2021, April). Visually grounded models of spoken language: A survey of datasets, architectures and evaluation techniques. Under review.
   Preprint
2. Chrupała, G., Higy, B., & Alishahi, A. (2020, July). Analyzing analytical methods: The case of phonology in neural models of spoken language. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (pp. 4146-4156).
   Paper | Code
3. Chrupała, G., & Alishahi, A. (2019, July). Correlating Neural and Symbolic Representations of Language. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics (pp. 2952-2962).
   Paper | Code
4. Alishahi, A., Barking, M., & Chrupała, G. (2017, August). Encoding of phonology in a recurrent neural model of grounded speech. In Proceedings of the 21st Conference on Computational Natural Language Learning (CoNLL 2017) (pp. 368-378). Best Paper Award.
   Paper | Code
5. Chrupała, G., Gelderloos, L., & Alishahi, A. (2017, July). Representations of language in a model of visually grounded speech signal. In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) (pp. 613-622).
   Paper | Code
6. Kádár, A., Chrupała, G., & Alishahi, A. (2017). Representation of linguistic form and function in recurrent neural networks. Computational Linguistics, 43(4), 761-780. Computational Linguistics, 43(4):761-780.
   Paper | Code

Recent Talks

• Visually Grounded Models of Spoken Language and their Analysis. Lecture at ALPS Winter School. Video
• Investigating neural representations of speech and language. Keynote at the Nordic Conference on Computational Linguistics (NoDaLiDa), October 2019. Slides

Supervision

Selected MSc theses

• Jeroen van der Weijden. 2019. Illuminating the black box : encoding of phonology in a recurrent neural model of grounded natural speech. Tilburg University.
• Kees Brekelmans. 2019. Evaluating Sentence Embeddings on Syntactic Information Through Representational Similarity Analysis. Tilburg University.
• Dennis de Groot. 2019. Finding Structure in Neural Network Activation Patterns via Representational Similarity and Convolutional Kernels. Tilburg University.
• Mark van der Laan. 2018. Encoding of speaker identity in a Neural Network model of Visually Grounded Speech perception. Tilburg University.
• Bart Broere. 2017. Syntactic properties of skip-thought vectors. Tiburg University.
• Lieke Gelderloos. 2016. Tilburg University. Levels of representation in a recurrent neural model of visually grounded language learning. Tilburg University. See also Coling 2016 paper: From phonemes to images: levels of representation in a recurrent neural model of visually grounded language learning.
• Kia Eisinga. 2016. Tilburg University. Predicting Runway Allocation with Support Vector Machine and Logistic Regression. Tilburg University.
• Ákos Kádár. 2014. Grounded learning for source code component retrieval. Tilburg University
• Antoaneta Baltadzhieva. 2014. Predicting question quality in question answering forums. Tilburg University
• Lucas Vergeest. 2014. Using N-grams and word embeddings for Twitter hashtag suggestion. Tilburg University
• Huijing Deng. 2013. Probabilistic Models of API Retrieval. Saarland University. (See also Deng and Chrupała. 2014. Semantic approaches to software component retrieval with English queries. LREC.)

Bio

Grzegorz Chrupała is an Associate Professor at the Department of Cognitive Science and Artificial Intelligence at Tilburg University. Previously he did postdoctoral research at the Spoken Language Systems group at Saarland University. He received his doctoral degree from the School of Computing at Dublin City University. His research focuses on computational models of language learning from multimodal signals such as speech and vision and on the analysis and interpretability of representations emerging in deep neural networks. He regularly serves as Area Chair for major NLP and AI conferences including ACL, NAACL, EMNLP and CoNLL. He co-organized the 2018, 2019 and 2020 editions of BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP. His research has been funded by the Dutch Research Council (NWO).

Contact

Twitter: @gchrupala
Web: grzegorz.chrupala.me
Email: grzegorz@chrupala.me