@inproceedings{wu-etal-2023-learning,
title = "Learning Action Conditions from Instructional Manuals for Instruction Understanding",
author = "Wu, Te-Lin and
Zhang, Caiqi and
Hu, Qingyuan and
Spangher, Alexander and
Peng, Nanyun",
editor = "Rogers, Anna and
Boyd-Graber, Jordan and
Okazaki, Naoaki",
booktitle = "Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
month = jul,
year = "2023",
address = "Toronto, Canada",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.acl-long.170/",
doi = "10.18653/v1/2023.acl-long.170",
pages = "3023--3043",
abstract = "The ability to infer pre- and postconditions of an action is vital for comprehending complex instructions, and is essential for applications such as autonomous instruction-guided agents and assistive AI that supports humans to perform physical tasks. In this work, we propose a task dubbed action condition inference, which extracts mentions of preconditions and postconditions of actions in instructional manuals. We propose a weakly supervised approach utilizing automatically constructed large-scale training instances from online instructions, and curate a densely human-annotated and validated dataset to study how well the current NLP models do on the proposed task. We design two types of models differ by whether contextualized and global information is leveraged, as well as various combinations of heuristics to construct the weak supervisions.Our experiments show a {\ensuremath{>}} 20{\%} F1-score improvement with considering the entire instruction contexts and a {\ensuremath{>}} 6{\%} F1-score benefit with the proposed heuristics. However, the best performing model is still well-behind human performance."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wu-etal-2023-learning">
<titleInfo>
<title>Learning Action Conditions from Instructional Manuals for Instruction Understanding</title>
</titleInfo>
<name type="personal">
<namePart type="given">Te-Lin</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Caiqi</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Qingyuan</namePart>
<namePart type="family">Hu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alexander</namePart>
<namePart type="family">Spangher</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nanyun</namePart>
<namePart type="family">Peng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Anna</namePart>
<namePart type="family">Rogers</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jordan</namePart>
<namePart type="family">Boyd-Graber</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Naoaki</namePart>
<namePart type="family">Okazaki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Toronto, Canada</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>The ability to infer pre- and postconditions of an action is vital for comprehending complex instructions, and is essential for applications such as autonomous instruction-guided agents and assistive AI that supports humans to perform physical tasks. In this work, we propose a task dubbed action condition inference, which extracts mentions of preconditions and postconditions of actions in instructional manuals. We propose a weakly supervised approach utilizing automatically constructed large-scale training instances from online instructions, and curate a densely human-annotated and validated dataset to study how well the current NLP models do on the proposed task. We design two types of models differ by whether contextualized and global information is leveraged, as well as various combinations of heuristics to construct the weak supervisions.Our experiments show a \ensuremath> 20% F1-score improvement with considering the entire instruction contexts and a \ensuremath> 6% F1-score benefit with the proposed heuristics. However, the best performing model is still well-behind human performance.</abstract>
<identifier type="citekey">wu-etal-2023-learning</identifier>
<identifier type="doi">10.18653/v1/2023.acl-long.170</identifier>
<location>
<url>https://aclanthology.org/2023.acl-long.170/</url>
</location>
<part>
<date>2023-07</date>
<extent unit="page">
<start>3023</start>
<end>3043</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Learning Action Conditions from Instructional Manuals for Instruction Understanding
%A Wu, Te-Lin
%A Zhang, Caiqi
%A Hu, Qingyuan
%A Spangher, Alexander
%A Peng, Nanyun
%Y Rogers, Anna
%Y Boyd-Graber, Jordan
%Y Okazaki, Naoaki
%S Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2023
%8 July
%I Association for Computational Linguistics
%C Toronto, Canada
%F wu-etal-2023-learning
%X The ability to infer pre- and postconditions of an action is vital for comprehending complex instructions, and is essential for applications such as autonomous instruction-guided agents and assistive AI that supports humans to perform physical tasks. In this work, we propose a task dubbed action condition inference, which extracts mentions of preconditions and postconditions of actions in instructional manuals. We propose a weakly supervised approach utilizing automatically constructed large-scale training instances from online instructions, and curate a densely human-annotated and validated dataset to study how well the current NLP models do on the proposed task. We design two types of models differ by whether contextualized and global information is leveraged, as well as various combinations of heuristics to construct the weak supervisions.Our experiments show a \ensuremath> 20% F1-score improvement with considering the entire instruction contexts and a \ensuremath> 6% F1-score benefit with the proposed heuristics. However, the best performing model is still well-behind human performance.
%R 10.18653/v1/2023.acl-long.170
%U https://aclanthology.org/2023.acl-long.170/
%U https://doi.org/10.18653/v1/2023.acl-long.170
%P 3023-3043
Markdown (Informal)
[Learning Action Conditions from Instructional Manuals for Instruction Understanding](https://aclanthology.org/2023.acl-long.170/) (Wu et al., ACL 2023)
ACL