Pith. sign in

REVIEW

Improving Compositional Generalization in Semantic Parsing

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2010.05647 v1 pith:YHIQKKB5 submitted 2020-10-12 cs.CL

Improving Compositional Generalization in Semantic Parsing

classification cs.CL
keywords generalizationcompositionalattentionsemanticdecoderimprovemodelsparsing
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Generalization of models to out-of-distribution (OOD) data has captured tremendous attention recently. Specifically, compositional generalization, i.e., whether a model generalizes to new structures built of components observed during training, has sparked substantial interest. In this work, we investigate compositional generalization in semantic parsing, a natural test-bed for compositional generalization, as output programs are constructed from sub-components. We analyze a wide variety of models and propose multiple extensions to the attention module of the semantic parser, aiming to improve compositional generalization. We find that the following factors improve compositional generalization: (a) using contextual representations, such as ELMo and BERT, (b) informing the decoder what input tokens have previously been attended to, (c) training the decoder attention to agree with pre-computed token alignments, and (d) downsampling examples corresponding to frequent program templates. While we substantially reduce the gap between in-distribution and OOD generalization, performance on OOD compositions is still substantially lower.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.