Explaining Translationese: why are Neural Classifiers Better and what do they Learn?

Kwabena Amponsah-Kaakyire, Daria Pylypenko, Josef Genabith, Cristina España-Bonet


Abstract
Recent work has shown that neural feature- and representation-learning, e.g. BERT, achieves superior performance over traditional manual feature engineering based approaches, with e.g. SVMs, in translationese classification tasks. Previous research did not show (i) whether the difference is because of the features, the classifiers or both, and (ii) what the neural classifiers actually learn. To address (i), we carefully design experiments that swap features between BERT- and SVM-based classifiers. We show that an SVM fed with BERT representations performs at the level of the best BERT classifiers, while BERT learning and using handcrafted features performs at the level of an SVM using handcrafted features. This shows that the performance differences are due to the features. To address (ii) we use integrated gradients and find that (a) there is indication that information captured by hand-crafted features is only a subset of what BERT learns, and (b) part of BERT’s top performance results are due to BERT learning topic differences and spurious correlations with translationese.
Anthology ID:
2022.blackboxnlp-1.23
Volume:
Proceedings of the Fifth BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP
Month:
December
Year:
2022
Address:
Abu Dhabi, United Arab Emirates (Hybrid)
Editors:
Jasmijn Bastings, Yonatan Belinkov, Yanai Elazar, Dieuwke Hupkes, Naomi Saphra, Sarah Wiegreffe
Venue:
BlackboxNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
281–296
Language:
URL:
https://aclanthology.org/2022.blackboxnlp-1.23
DOI:
10.18653/v1/2022.blackboxnlp-1.23
Bibkey:
Cite (ACL):
Kwabena Amponsah-Kaakyire, Daria Pylypenko, Josef Genabith, and Cristina España-Bonet. 2022. Explaining Translationese: why are Neural Classifiers Better and what do they Learn?. In Proceedings of the Fifth BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP, pages 281–296, Abu Dhabi, United Arab Emirates (Hybrid). Association for Computational Linguistics.
Cite (Informal):
Explaining Translationese: why are Neural Classifiers Better and what do they Learn? (Amponsah-Kaakyire et al., BlackboxNLP 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.blackboxnlp-1.23.pdf