Grzegorz Chrupała

Dr. Grzegorz Chrupała

orcid

Research Focus & Metrics

Academic Career Length

24Years (2003-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 December 2017 in Computational Linguistics
Verified
Journal Article
Open access
ÁKÁkos Kádár
Ákos Kádár
GCGrzegorz Chrupała
Grzegorz Chrupała
AAAfra Alishahi
Afra Alishahi
We present novel methods for analyzing the activation patterns of recurrent neural networks from a linguistic point of view and explore the types of linguistic structure they learn. As a case study, we use a standard standalone language model, and a multi-task gated recurrent network architecture consisting of two parallel pathways with shared word embeddings: The Visual pathway is trained on predicting the representations of the visual scene corresponding to an input sentence, and the Textual pathway is trained to predict the next word in the same sentence. We propose a method for estimating the amount of contribution of individual tokens in the input to the final prediction of the networks. Using this method, we show that the Visual pathway pays selective attention to lexical categories and grammatical functions that carry semantic information, and learns to treat word types differently depending on their grammatical function and their position in the sequential structure of the sentence. In contrast, the language models are comparatively more sensitive to words with a syntactic function. Further analysis of the most informative n-gram contexts for each model shows that in comparison with the Visual pathway, the language models react more strongly to abstract contexts that represent syntactic constructions.

1 January 2017
Verified
Conference Paper
Open access
GCGrzegorz Chrupała
Grzegorz Chrupała
LGLieke Gelderloos
Lieke Gelderloos
AAAfra Alishahi
Afra Alishahi

31 July 2019 in Natural Language Engineering
Verified
Journal Article
Closed access
PDF file
AAAfra Alishahi
Afra Alishahi
GCGrzegorz Chrupała
Grzegorz Chrupała
TLTal Linzen
Tal Linzen
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.

1 January 2019
Verified
Conference Paper
Open access
GCGrzegorz Chrupała
Grzegorz Chrupała
AAAfra Alishahi
Afra Alishahi

29 September 2015 in ACM SIGKDD Explorations Newsletter
Verified
Journal Article
Open access
PDF file
ABAntoaneta Baltadzhieva
Antoaneta Baltadzhieva
GCGrzegorz Chrupała
Grzegorz Chrupała
Community Question Answering websites (CQA) offer a new opportunity for users to provide, search and share knowledge. Although the idea of receiving a direct, targeted response to a question sounds very attractive, the quality of the question itself can have an important effect on the likelihood of getting useful answers. High quality questions improve the CQA experience and therefore it is essential for CQA forums to better understand what characterizes questions that are more appealing for the forum community. In this survey, we review existing research on question quality in CQA websites. We discuss the possible measures of question quality and the question features that have been shown to influence question quality.

1 January 2017
Verified
Conference Paper
Open access
AAAfra Alishahi
Afra Alishahi
MBMarie Barking
Marie Barking
GCGrzegorz Chrupała
Grzegorz Chrupała