CASE: Commonsense-Augmented Score with an Expanded Answer Space

Wenkai Chen, Sahithya Ravi, Vered Shwartz

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

LLMs have demonstrated impressive zero-shot performance on NLP tasks thanks to the knowledge they acquired in their training. In multiple-choice QA tasks, the LM probabilities are used as an imperfect measure of the plausibility of each answer choice. One of the major limitations of the basic score is that it treats all words as equally important. We propose CASE, a Commonsense-Augmented Score with an Expanded Answer Space. CASE addresses this limitation by assigning importance weights for individual words based on their semantic relations to other words in the input. The dynamic weighting approach outperforms basic LM scores, not only because it reduces noise from unimportant words, but also because it informs the model of implicit commonsense knowledge that may be useful for answering the question. We then also follow prior work in expanding the answer space by generating lexically-divergent answers that are conceptually-similar to the choices. When combined with answer space expansion, our method outperforms strong baselines on 5 commonsense benchmarks. We further show these two approaches are complementary and may be especially beneficial when using smaller LMs.

Original languageEnglish
Title of host publicationFindings of the Association for Computational Linguistics
Subtitle of host publicationEMNLP 2023
PublisherAssociation for Computational Linguistics (ACL)
Pages2732-2744
Number of pages13
ISBN (Electronic)9798891760615
DOIs
StatePublished - 2023
Externally publishedYes
Event2023 Findings of the Association for Computational Linguistics: EMNLP 2023 - Hybrid, Singapore
Duration: 6 Dec 202310 Dec 2023

Publication series

NameFindings of the Association for Computational Linguistics: EMNLP 2023

Conference

Conference2023 Findings of the Association for Computational Linguistics: EMNLP 2023
Country/TerritorySingapore
CityHybrid
Period6/12/2310/12/23

Bibliographical note

Publisher Copyright:
© 2023 Association for Computational Linguistics.

Funding

This work was funded, in part, by the Vector Institute for AI, Canada CIFAR AI Chairs program, an NSERC discovery grant, and a research gift from AI2.

FundersFunder number
Natural Sciences and Engineering Research Council of Canada
Vector Institute

    Fingerprint

    Dive into the research topics of 'CASE: Commonsense-Augmented Score with an Expanded Answer Space'. Together they form a unique fingerprint.

    Cite this