Adverbs, Surprisingly

Dmitry Nikolaev; Collin F. Baker; Miriam R. L. Petruck; Sebastian Padó

doi:10.18653/v1/2023.starsem-1.44

Adverbs, Surprisingly

Dmitry Nikolaev, Collin F. Baker, Miriam R. L. Petruck, Sebastian Padó

University of Stuttgart

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

39 Downloads (Pure)

Abstract

This paper begins with the premise that adverbs are neglected in computational linguistics. This view derives from two analyses: a literature review and a novel adverb dataset to probe a state-of-the-art language model, thereby uncovering systematic gaps in accounts for adverb meaning. We suggest that using Frame Semantics for characterizing word meaning, as in FrameNet, provides a promising approach to adverb analysis, given its ability to describe ambiguity, semantic roles, and null instantiation.

Original language	English
Title of host publication	Proceedings of the 12th Joint Conference on Lexical and Computational Semantics (*SEM 2023)
Place of Publication	Toronto, Canada
Publisher	Association for Computational Linguistics
Pages	512
Number of pages	526
DOIs	https://doi.org/10.18653/v1/2023.starsem-1.44
Publication status	Published - Jul 2023
Externally published	Yes
Event	The 12th Joint Conference on Lexical and Computational Semantics (*SEM 2023) - Toronto, Canada Duration: 13 Jul 2023 → 14 Jul 2023 https://sites.google.com/view/starsem2023

Conference

Conference	The 12th Joint Conference on Lexical and Computational Semantics (*SEM 2023)
Country/Territory	Canada
City	Toronto
Period	13/07/23 → 14/07/23
Internet address	https://sites.google.com/view/starsem2023

Keywords

cs.CL

Access to Document

10.18653/v1/2023.starsem-1.44

2305.19650v1

Cite this

@inproceedings{b7632e786e8d49089cffc8a6a9228add,

title = "Adverbs, Surprisingly",

abstract = "This paper begins with the premise that adverbs are neglected in computational linguistics. This view derives from two analyses: a literature review and a novel adverb dataset to probe a state-of-the-art language model, thereby uncovering systematic gaps in accounts for adverb meaning. We suggest that using Frame Semantics for characterizing word meaning, as in FrameNet, provides a promising approach to adverb analysis, given its ability to describe ambiguity, semantic roles, and null instantiation. ",

keywords = "cs.CL",

author = "Dmitry Nikolaev and Baker, {Collin F.} and Petruck, {Miriam R. L.} and Sebastian Pad{\'o}",

year = "2023",

month = jul,

doi = "10.18653/v1/2023.starsem-1.44",

language = "English",

pages = "512",

booktitle = "Proceedings of the 12th Joint Conference on Lexical and Computational Semantics (*SEM 2023)",

publisher = "Association for Computational Linguistics",

address = "United States",

note = "The 12th Joint Conference on Lexical and Computational Semantics (*SEM 2023) ; Conference date: 13-07-2023 Through 14-07-2023",

url = "https://sites.google.com/view/starsem2023",

}

Nikolaev, D, Baker, CF, Petruck, MRL & Padó, S 2023, Adverbs, Surprisingly. in Proceedings of the 12th Joint Conference on Lexical and Computational Semantics (*SEM 2023). Association for Computational Linguistics, Toronto, Canada, pp. 512, The 12th Joint Conference on Lexical and Computational Semantics (*SEM 2023), Toronto, Canada, 13/07/23. https://doi.org/10.18653/v1/2023.starsem-1.44

TY - GEN

T1 - Adverbs, Surprisingly

AU - Nikolaev, Dmitry

AU - Baker, Collin F.

AU - Petruck, Miriam R. L.

AU - Padó, Sebastian

PY - 2023/7

Y1 - 2023/7

N2 - This paper begins with the premise that adverbs are neglected in computational linguistics. This view derives from two analyses: a literature review and a novel adverb dataset to probe a state-of-the-art language model, thereby uncovering systematic gaps in accounts for adverb meaning. We suggest that using Frame Semantics for characterizing word meaning, as in FrameNet, provides a promising approach to adverb analysis, given its ability to describe ambiguity, semantic roles, and null instantiation.

AB - This paper begins with the premise that adverbs are neglected in computational linguistics. This view derives from two analyses: a literature review and a novel adverb dataset to probe a state-of-the-art language model, thereby uncovering systematic gaps in accounts for adverb meaning. We suggest that using Frame Semantics for characterizing word meaning, as in FrameNet, provides a promising approach to adverb analysis, given its ability to describe ambiguity, semantic roles, and null instantiation.

KW - cs.CL

U2 - 10.18653/v1/2023.starsem-1.44

DO - 10.18653/v1/2023.starsem-1.44

M3 - Conference contribution

SP - 512

BT - Proceedings of the 12th Joint Conference on Lexical and Computational Semantics (*SEM 2023)

PB - Association for Computational Linguistics

CY - Toronto, Canada

T2 - The 12th Joint Conference on Lexical and Computational Semantics (*SEM 2023)

Y2 - 13 July 2023 through 14 July 2023

ER -

Adverbs, Surprisingly

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this