We present a methodology that explores how sentence structure is reflected in neural representations of machine translation systems. We demonstrate our model-agnostic approach with the Transformer English-German translation model. We analyze neuron-level correlation of activations between paraphrases while discussing the methodology challenges and the need for confound analysis to isolate the effects of shallow cues. We find that similarity between activation patterns can be mostly accounted for by similarity in word choice and sentence length. Following that, we manipulate neuron activations to control the syntactic form of the output. We show this intervention to be somewhat successful, indicating that deep models capture sentence-structure distinctions, despite finding no such indication at the neuron level. To conduct our experiments, we develop a semi-automatic method to generate meaning-preserving minimal pair paraphrases (active-passive voice and adverbial clause-noun phrase) and compile a corpus of such pairs.
@inproceedings{DBLP:conf/conll/PatelCA22,
author = {Gal Patel and
Leshem Choshen and
Omri Abend},
editor = {Antske Fokkens and
Vivek Srikumar},
title = {On Neurons Invariant to Sentence Structural Changes in Neural Machine
Translation},
booktitle = {Proceedings of the 26th Conference on Computational Natural Language
Learning, CoNLL 2022, Abu Dhabi, United Arab Emirates (Hybrid Event),
December 7-8, 2022},
pages = {194--212},
publisher = {Association for Computational Linguistics},
year = {2022},
url = {https://doi.org/10.18653/v1/2022.conll-1.14},
doi = {10.18653/V1/2022.CONLL-1.14},
timestamp = {Fri, 12 Apr 2024 01:00:00 +0200},
biburl = {https://dblp.org/rec/conf/conll/PatelCA22.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
See the full reference list in the paper.