The 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Proceedings of the Conference)
Association for Computational Linguistics
Recent work has shown that LSTMs trained on a generic language modeling objective capture syntax-sensitive generalizations such as long-distance number agreement. We have however no mechanistic understanding of how they accomplish this remarkable feat. Some have conjectured it depends on heuristics that do not truly take hierarchical structure into account. We present here a detailed study of the inner mechanics of number tracking in LSTMs at the single neuron level. We discover that long-distance number information is largely managed by two ¿number units¿. Importantly, the behaviour of these units is partially controlled by other units independently shown to track syntactic structure. We conclude that LSTMs are, to some extent, implementing genuinely syntactic processing mechanisms, paving the way to a more general understanding of grammatical encoding in LSTMs.
Lakretz, Y.; Kruszewski, G.; Desbordes, T.; Hupkes, D.; Dehaene, S.; Baroni, M.. The emergence of number and syntax units in LSTM language models. In: Burstein, J.; Doran, C.; Solorio, T. (eds.). The 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Proceedings of the Conference). 1 ed. East Stroudsburg PA: Association for Computational Linguistics; 2019. p. 11-20.