Latent compositional representations improve systematic generalization in grounded question answering

Ben Bogin, Sanjay Subramanian, Matt Gardner, Jonathan Berant

Research output: Contribution to journalArticlepeer-review

15 Scopus citations

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 CLOSURE, 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.

Original languageEnglish
Pages (from-to)195-210
Number of pages16
JournalTransactions of the Association for Computational Linguistics
Volume9
DOIs
StatePublished - 1 Feb 2021

Funding

FundersFunder number
European Union Horizons 2020 research and innovation programmeDELPHI 802800
Jonathan Herzig
Yandex Initiative for Machine Learning
Horizon 2020 Framework Programme802800
Horizon 2020 Framework Programme
European Research Council

    Fingerprint

    Dive into the research topics of 'Latent compositional representations improve systematic generalization in grounded question answering'. Together they form a unique fingerprint.

    Cite this