@inproceedings{miller-2026-documenting,
title = "Documenting Corporate Harm: A Semantic Action Trajectories Approach to the Opioid Industry Document Archive Shared Task",
author = "Miller, Ben",
editor = "Card, Dallas and
Field, Anjalie and
Keith, Katherine and
Mendelsohn, Julia",
booktitle = "Proceedings of the Seventh Workshop on Natural Language Processing and Computational Social Science",
month = jul,
year = "2026",
address = "San Diego",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.nlpcss-1.10/",
pages = "149--158",
ISBN = "979-8-89176-426-2",
abstract = "This paper presents a method for modeling change in the possibility space of actors over time as represented in the Opioid Industry Document Archive (OIDA). The approach treats documents as a structured field of actor{--}action relations and models these relations as semantic action trajectories across time. Semantic role labeling (SRL) using the Emory Language and Information Toolkit (ELIT) is applied to extract subject{--}predicate structures from a corpus of internal industry documents. Subjects are normalized and grouped into actor categories using a combination of rule-based heuristics and constrained language model adjudication. Predicate vocabularies associated with these actors are mapped to psycholinguistic categories using the LIWC lexicon, and random forest feature selection with principal component analysis is used to construct a low-dimensional representation of discourse structure across periods.The resulting discourse space reveals systematic shifts in how corporate actors, regulators, clinicians, and patients are positioned over time. In particular, corporate entities and the opioid products they produce follow nearly identical semantic trajectories, suggesting that companies and the pharmaceutical drugs they produce occupy similar roles in the archive{'}s discourse. This method provides a way to analyze changing institutional behavior at scale across heterogeneous litigation and historical archives."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="miller-2026-documenting">
<titleInfo>
<title>Documenting Corporate Harm: A Semantic Action Trajectories Approach to the Opioid Industry Document Archive Shared Task</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ben</namePart>
<namePart type="family">Miller</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Seventh Workshop on Natural Language Processing and Computational Social Science</title>
</titleInfo>
<name type="personal">
<namePart type="given">Dallas</namePart>
<namePart type="family">Card</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Anjalie</namePart>
<namePart type="family">Field</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Katherine</namePart>
<namePart type="family">Keith</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Julia</namePart>
<namePart type="family">Mendelsohn</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">San Diego</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-426-2</identifier>
</relatedItem>
<abstract>This paper presents a method for modeling change in the possibility space of actors over time as represented in the Opioid Industry Document Archive (OIDA). The approach treats documents as a structured field of actor–action relations and models these relations as semantic action trajectories across time. Semantic role labeling (SRL) using the Emory Language and Information Toolkit (ELIT) is applied to extract subject–predicate structures from a corpus of internal industry documents. Subjects are normalized and grouped into actor categories using a combination of rule-based heuristics and constrained language model adjudication. Predicate vocabularies associated with these actors are mapped to psycholinguistic categories using the LIWC lexicon, and random forest feature selection with principal component analysis is used to construct a low-dimensional representation of discourse structure across periods.The resulting discourse space reveals systematic shifts in how corporate actors, regulators, clinicians, and patients are positioned over time. In particular, corporate entities and the opioid products they produce follow nearly identical semantic trajectories, suggesting that companies and the pharmaceutical drugs they produce occupy similar roles in the archive’s discourse. This method provides a way to analyze changing institutional behavior at scale across heterogeneous litigation and historical archives.</abstract>
<identifier type="citekey">miller-2026-documenting</identifier>
<location>
<url>https://aclanthology.org/2026.nlpcss-1.10/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>149</start>
<end>158</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Documenting Corporate Harm: A Semantic Action Trajectories Approach to the Opioid Industry Document Archive Shared Task
%A Miller, Ben
%Y Card, Dallas
%Y Field, Anjalie
%Y Keith, Katherine
%Y Mendelsohn, Julia
%S Proceedings of the Seventh Workshop on Natural Language Processing and Computational Social Science
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego
%@ 979-8-89176-426-2
%F miller-2026-documenting
%X This paper presents a method for modeling change in the possibility space of actors over time as represented in the Opioid Industry Document Archive (OIDA). The approach treats documents as a structured field of actor–action relations and models these relations as semantic action trajectories across time. Semantic role labeling (SRL) using the Emory Language and Information Toolkit (ELIT) is applied to extract subject–predicate structures from a corpus of internal industry documents. Subjects are normalized and grouped into actor categories using a combination of rule-based heuristics and constrained language model adjudication. Predicate vocabularies associated with these actors are mapped to psycholinguistic categories using the LIWC lexicon, and random forest feature selection with principal component analysis is used to construct a low-dimensional representation of discourse structure across periods.The resulting discourse space reveals systematic shifts in how corporate actors, regulators, clinicians, and patients are positioned over time. In particular, corporate entities and the opioid products they produce follow nearly identical semantic trajectories, suggesting that companies and the pharmaceutical drugs they produce occupy similar roles in the archive’s discourse. This method provides a way to analyze changing institutional behavior at scale across heterogeneous litigation and historical archives.
%U https://aclanthology.org/2026.nlpcss-1.10/
%P 149-158
Markdown (Informal)
[Documenting Corporate Harm: A Semantic Action Trajectories Approach to the Opioid Industry Document Archive Shared Task](https://aclanthology.org/2026.nlpcss-1.10/) (Miller, NLP+CSS 2026)
ACL