Abstract
Since their inception, encoder-decoder models have successfully been applied to a wide array of problems in computational linguistics. The most recent successes are predominantly due to the use of different variations of attention mechanisms, but their cognitive plausibility is questionable. In particular, because past representations can be revisited at any point in time, attention-centric methods seem to lack an incentive to build up incrementally more informative representations of incoming sentences. This way of processing stands in stark contrast with the way in which humans are believed to process language: continuously and rapidly integrating new information as it is encountered. In this work, we propose three novel metrics to assess the behavior of RNNs with and without an attention mechanism and identify key differences in the way the different model types process sentences.- Anthology ID:
- W19-4324
- Volume:
- Proceedings of the 4th Workshop on Representation Learning for NLP (RepL4NLP-2019)
- Month:
- August
- Year:
- 2019
- Address:
- Florence, Italy
- Venue:
- RepL4NLP
- SIG:
- SIGREP
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 209–217
- Language:
- URL:
- https://aclanthology.org/W19-4324
- DOI:
- 10.18653/v1/W19-4324
- Cite (ACL):
- Dennis Ulmer, Dieuwke Hupkes, and Elia Bruni. 2019. Assessing Incrementality in Sequence-to-Sequence Models. In Proceedings of the 4th Workshop on Representation Learning for NLP (RepL4NLP-2019), pages 209–217, Florence, Italy. Association for Computational Linguistics.
- Cite (Informal):
- Assessing Incrementality in Sequence-to-Sequence Models (Ulmer et al., RepL4NLP 2019)
- PDF:
- https://preview.aclanthology.org/remove-xml-comments/W19-4324.pdf
- Code
- i-machine-think/incremental_encoding
- Data
- SCAN