Latent Compositional Representations Improve Systematic Generalization in Grounded Question Answering
Ben Bogin, Sanjay Subramanian, Matt Gardner, Jonathan Berant
Abstract
Answering questions that involve multi-step reasoning requires decomposing them and using the answers of intermediate steps to reach the final answer. However, state-of-the-art models in grounded question answering often do not explicitly perform decomposition, leading to difficulties in generalization to out-of-distribution examples. In this work, we propose a model that computes a representation and denotation for all question spans in a bottom-up, compositional manner using a CKY-style parser. Our model induces latent trees, driven by end-to-end (the answer) supervision only. We show that this inductive bias towards tree structures dramatically improves systematic generalization to out-of- distribution examples, compared to strong baselines on an arithmetic expressions benchmark as well as on C losure, a dataset that focuses on systematic generalization for grounded question answering. On this challenging dataset, our model reaches an accuracy of 96.1%, significantly higher than prior models that almost perfectly solve the task on a random, in-distribution split.- Anthology ID:
- 2021.tacl-1.12
- Volume:
- Transactions of the Association for Computational Linguistics, Volume 9
- Month:
- Year:
- 2021
- Address:
- Cambridge, MA
- Editors:
- Brian Roark, Ani Nenkova
- Venue:
- TACL
- SIG:
- Publisher:
- MIT Press
- Note:
- Pages:
- 195–210
- Language:
- URL:
- https://preview.aclanthology.org/build-pipeline-with-new-library/2021.tacl-1.12/
- DOI:
- 10.1162/tacl_a_00361
- Cite (ACL):
- Ben Bogin, Sanjay Subramanian, Matt Gardner, and Jonathan Berant. 2021. Latent Compositional Representations Improve Systematic Generalization in Grounded Question Answering. Transactions of the Association for Computational Linguistics, 9:195–210.
- Cite (Informal):
- Latent Compositional Representations Improve Systematic Generalization in Grounded Question Answering (Bogin et al., TACL 2021)
- PDF:
- https://preview.aclanthology.org/build-pipeline-with-new-library/2021.tacl-1.12.pdf