About myself
I study cognitive semantic approaches to the modeling of causal structure in language for AI interpretability and reasoning. One focus of my work is creating and curating natural language resources for model evaluation across a variety of AI and NLP tasks including causal story induction, physical commonsense reasoning, knowledge base relation hallucination detection, and multilingual QA. Graphs are everywhere, and a key question in my research is whether semantic structures (e.g., causal graphs or other symbolic meaning representations) can help increase the interpretability of AI systems. My approach to language study adopts perspectives from functional and cognitive linguistics, where a central hypothesis is that participant force-dynamic interactions in the physical world (e.g., an event like 'I dropped the ball') shape both human cognition (something akin to an intuitive physics or other abstract knowledge representation) and also linguistic constructions (grammatical categories associated with complex syntactic form). As such, we can study grammatical form associated with causal meaning and, ultimately, gain insight into human and machine reasoning processes. I was a postdoc at the University of Washington working with Yejin Choi, I have a PhD in Linguistics for research on causal structure discovery, an MSc in Computer Science with emphasis on machine learning and experimental methods, and did internships at Amazon (Seattle and Boston) doing work on entity reference resolution and at Sandia National Laboratories building tools for ontology induction from document collections.
I led a Computational Linguistics graduate seminar at the University of Washington, Winter 2024!
- Postdoctoral Researcher, University of Washington, Paul G. Allen School of Computer Science & Engineering
- Causal schemas; Datasets for causal reasoning; Meaning representations for hallucination detection; Narrative structure understanding.
Invited talks
- Jun 2023: University of Zurich, Multicast, Dept. of Psychology, hosted by Prof. Dr. Birgit Kleim
- Nov 2019: UIUC Blender Lab NLP Seminar, hosted by Prof. Heng Ji
- Nov 2019: CU Boulder CompSem
- Oct 2019: Amazon (Boston) Tech Talk
- Oct 2019: Brandeis Computational Linguistics
Selected Publications
Causal structure for schema and narrative understanding
-
Causal schema induction for knowledge discovery
[Michael Regan, Jena Hwang, Keisuke Sakaguchi, and James Pustejovsky]
Paper
Code
Data
-
RESIN-11: Schema-guided Event Prediction for 11 Newsworthy Scenarios
[Xinya Du, ..., Michael Regan, ..., Heng Ji]
Demo track, NAACL 2022.
Paper
-
[Michael Regan, James Pustejovsky, and William Croft]
Extraction of causal structure from procedural text for discourse representations
AKBC, SciNLP Workshop, 2020.
-
Decomposing Events and Storylines
[William Croft, Pavlina Kalm, and Michael Regan]
In: Computational Analysis of Storylines: Making Sense of Events, Cambridge University Press, 2021.
Discourse, dialogue, and utterance meaning representations
-
MASSIVE Multilingual Abstract Meaning Representation: A Dataset and Baselines for Hallucination Detection
[Michael Regan, Shira Wein, George Baker, and Emilio Monti]
Accepted (oral presentation): *SEM 2024 (13th Joint Conference on Lexical and Computational Semantics).
Paper
Sample code!
Data
-
Linear Cross-document Event Coreference Resolution using X-AMR
[Shafiuddin Rehan Ahmed, Evi Judge, George Baker, Michael Regan, Martha Palmer, and James H. Martin]
LREC-COLING, 2024.
Paper
-
How Good is the Model in Model-in-the-loop Event Coreference Resolution Annotation?
[Shafiuddin Rehan Ahmed, Abhijnan Nath, Michael Regan, Adam Pollins, Nikhil Krishnaswamy, James H. Martin]
Linguistics Annotation Workshop (LAW-XVII), ACL, 2023.
Paper
Code
Data
-
A Comparative Analysis of Automatic Speech Recognition Errors in Small Group Classroom Discourse
[Jie Cao, Ananya Ganesh, Jon Cai, Rosy Southwell, E. Margaret Perkoff, Michael Regan, Katharina Kann, James Martin, Martha Palmer, and Sidney D'Mello]
ACM User Modeling, Adaptation and Personalization (UMAP), 2023.
-
Representing constructional metaphors
[Pavlina Kalm, Michael Regan, Sook-kyung Lee, Chris Peverada and William Croft]
Second International Workshop on Designing Meaning Representations, COLING 2020.
-
Event structure representation: Between verbs and argument structure constructions
[Pavlina Kalm, Michael Regan, William Croft]
First International Workshop on Designing Meaning Representations, ACL 2019.
-
CRAFT Shared Tasks 2019 Overview -- Integrated Structure, Semantics, and Coreference
[William Baumgartner, ..., Michael Regan and Lawrence Hunter]
5th Workshop on BioNLP Open Shared Tasks, EMNLP 2019.
-
Tense and aspect semantics for sentential AMR
[Lucia Donatelli, Michael Regan, William Croft, and Nathan Schneider]
SCiL, 2019.
-
A dataset for resolving referring expressions in spoken dialogue via contextual query rewrites (CQR).
[Michael Regan, Pushpendre Rastogi, Arpit Gupta, and Lambert Mathias]
arXiv, 2019.
Paper
Data
-
AMR beyond the sentence: the multi-sentence AMR corpus.
[Tim O'Gorman, Michael Regan, Kira Griffitt, Ulf Hermjakob, Kevin Knight, and Martha Palmer]
COLING, 2018.
Semantic typology
-
Cross-linguistic semantic annotation: Reconciling the language-specific and the universal.
[Jens Van Gysel, Meagan Vigus, Pavlina Kalm, Sook-kyung Lee, Michael Regan, and William Croft]
First International Workshop on Designing Meaning Representations, ACL 2019.
-
Linguistic typology meets Universal Dependencies.
[William Croft, Dawn Nordquist, Katherine Looney, and Michael Regan]
TLT 2017: 15th International Workshop on Treebanks and Linguistic Theories.
Service
- Reviewer: COLING 2020, 2022, 2024; NAACL 2021, 2022; GURT 2023; ACL 2023; *SEM 2023, 2024
- Program committee: SigTyp 2020, 2021, 2023
- I study cognitive semantic approaches to the modeling of causal structure in language for AI interpretability and reasoning. One focus of my work is creating and curating natural language resources for model evaluation across a variety of AI and NLP tasks including causal story induction, physical commonsense reasoning, knowledge base relation hallucination detection, and multilingual QA. Graphs are everywhere, and a key question in my research is whether semantic structures (e.g., causal graphs or other symbolic meaning representations) can help increase the interpretability of AI systems. My approach to language study adopts perspectives from functional and cognitive linguistics, where a central hypothesis is that participant force-dynamic interactions in the physical world (e.g., an event like 'I dropped the ball') shape both human cognition (something akin to an intuitive physics or other abstract knowledge representation) and also linguistic constructions (grammatical categories associated with complex syntactic form). As such, we can study grammatical form associated with causal meaning and, ultimately, gain insight into human and machine reasoning processes. I was a postdoc at the University of Washington working with Yejin Choi, I have a PhD in Linguistics for research on causal structure discovery, an MSc in Computer Science with emphasis on machine learning and experimental methods, and did internships at Amazon (Seattle and Boston) doing work on entity reference resolution and at Sandia National Laboratories building tools for ontology induction from document collections.
- I led a Computational Linguistics graduate seminar at the University of Washington, Winter 2024!
- Postdoctoral Researcher, University of Washington, Paul G. Allen School of Computer Science & Engineering
- Causal schemas; Datasets for causal reasoning; Meaning representations for hallucination detection; Narrative structure understanding.
- Jun 2023: University of Zurich, Multicast, Dept. of Psychology, hosted by Prof. Dr. Birgit Kleim
- Nov 2019: UIUC Blender Lab NLP Seminar, hosted by Prof. Heng Ji
- Nov 2019: CU Boulder CompSem
- Oct 2019: Amazon (Boston) Tech Talk
- Oct 2019: Brandeis Computational Linguistics
Selected Publications
Causal structure for schema and narrative understanding
-
Causal schema induction for knowledge discovery
[Michael Regan, Jena Hwang, Keisuke Sakaguchi, and James Pustejovsky]
Paper
Code
Data
-
RESIN-11: Schema-guided Event Prediction for 11 Newsworthy Scenarios
[Xinya Du, ..., Michael Regan, ..., Heng Ji]
Demo track, NAACL 2022.
Paper
-
[Michael Regan, James Pustejovsky, and William Croft]
Extraction of causal structure from procedural text for discourse representations
AKBC, SciNLP Workshop, 2020.
-
Decomposing Events and Storylines
[William Croft, Pavlina Kalm, and Michael Regan]
In: Computational Analysis of Storylines: Making Sense of Events, Cambridge University Press, 2021.
Discourse, dialogue, and utterance meaning representations
-
MASSIVE Multilingual Abstract Meaning Representation: A Dataset and Baselines for Hallucination Detection
[Michael Regan, Shira Wein, George Baker, and Emilio Monti]
Accepted (oral presentation): *SEM 2024 (13th Joint Conference on Lexical and Computational Semantics).
Paper
Sample code!
Data
-
Linear Cross-document Event Coreference Resolution using X-AMR
[Shafiuddin Rehan Ahmed, Evi Judge, George Baker, Michael Regan, Martha Palmer, and James H. Martin]
LREC-COLING, 2024.
Paper
-
How Good is the Model in Model-in-the-loop Event Coreference Resolution Annotation?
[Shafiuddin Rehan Ahmed, Abhijnan Nath, Michael Regan, Adam Pollins, Nikhil Krishnaswamy, James H. Martin]
Linguistics Annotation Workshop (LAW-XVII), ACL, 2023.
Paper
Code
Data
-
A Comparative Analysis of Automatic Speech Recognition Errors in Small Group Classroom Discourse
[Jie Cao, Ananya Ganesh, Jon Cai, Rosy Southwell, E. Margaret Perkoff, Michael Regan, Katharina Kann, James Martin, Martha Palmer, and Sidney D'Mello]
ACM User Modeling, Adaptation and Personalization (UMAP), 2023.
-
Representing constructional metaphors
[Pavlina Kalm, Michael Regan, Sook-kyung Lee, Chris Peverada and William Croft]
Second International Workshop on Designing Meaning Representations, COLING 2020.
-
Event structure representation: Between verbs and argument structure constructions
[Pavlina Kalm, Michael Regan, William Croft]
First International Workshop on Designing Meaning Representations, ACL 2019.
-
CRAFT Shared Tasks 2019 Overview -- Integrated Structure, Semantics, and Coreference
[William Baumgartner, ..., Michael Regan and Lawrence Hunter]
5th Workshop on BioNLP Open Shared Tasks, EMNLP 2019.
-
Tense and aspect semantics for sentential AMR
[Lucia Donatelli, Michael Regan, William Croft, and Nathan Schneider]
SCiL, 2019.
-
A dataset for resolving referring expressions in spoken dialogue via contextual query rewrites (CQR).
[Michael Regan, Pushpendre Rastogi, Arpit Gupta, and Lambert Mathias]
arXiv, 2019.
Paper
Data
-
AMR beyond the sentence: the multi-sentence AMR corpus.
[Tim O'Gorman, Michael Regan, Kira Griffitt, Ulf Hermjakob, Kevin Knight, and Martha Palmer]
COLING, 2018.
Semantic typology
-
Cross-linguistic semantic annotation: Reconciling the language-specific and the universal.
[Jens Van Gysel, Meagan Vigus, Pavlina Kalm, Sook-kyung Lee, Michael Regan, and William Croft]
First International Workshop on Designing Meaning Representations, ACL 2019.
-
Linguistic typology meets Universal Dependencies.
[William Croft, Dawn Nordquist, Katherine Looney, and Michael Regan]
TLT 2017: 15th International Workshop on Treebanks and Linguistic Theories.
Service
- Reviewer: COLING 2020, 2022, 2024; NAACL 2021, 2022; GURT 2023; ACL 2023; *SEM 2023, 2024
- Program committee: SigTyp 2020, 2021, 2023
Causal structure for schema and narrative understanding
-
Causal schema induction for knowledge discovery
[Michael Regan, Jena Hwang, Keisuke Sakaguchi, and James Pustejovsky]
Paper Code Data -
RESIN-11: Schema-guided Event Prediction for 11 Newsworthy Scenarios
[Xinya Du, ..., Michael Regan, ..., Heng Ji]
Demo track, NAACL 2022.
Paper -
[Michael Regan, James Pustejovsky, and William Croft]
Extraction of causal structure from procedural text for discourse representations
AKBC, SciNLP Workshop, 2020. -
Decomposing Events and Storylines
[William Croft, Pavlina Kalm, and Michael Regan]
In: Computational Analysis of Storylines: Making Sense of Events, Cambridge University Press, 2021.
Discourse, dialogue, and utterance meaning representations
-
MASSIVE Multilingual Abstract Meaning Representation: A Dataset and Baselines for Hallucination Detection
[Michael Regan, Shira Wein, George Baker, and Emilio Monti]
Accepted (oral presentation): *SEM 2024 (13th Joint Conference on Lexical and Computational Semantics).
Paper Sample code! Data -
Linear Cross-document Event Coreference Resolution using X-AMR
[Shafiuddin Rehan Ahmed, Evi Judge, George Baker, Michael Regan, Martha Palmer, and James H. Martin]
LREC-COLING, 2024.
Paper -
How Good is the Model in Model-in-the-loop Event Coreference Resolution Annotation?
[Shafiuddin Rehan Ahmed, Abhijnan Nath, Michael Regan, Adam Pollins, Nikhil Krishnaswamy, James H. Martin]
Linguistics Annotation Workshop (LAW-XVII), ACL, 2023.
Paper Code Data -
A Comparative Analysis of Automatic Speech Recognition Errors in Small Group Classroom Discourse
[Jie Cao, Ananya Ganesh, Jon Cai, Rosy Southwell, E. Margaret Perkoff, Michael Regan, Katharina Kann, James Martin, Martha Palmer, and Sidney D'Mello]
ACM User Modeling, Adaptation and Personalization (UMAP), 2023. -
Representing constructional metaphors
[Pavlina Kalm, Michael Regan, Sook-kyung Lee, Chris Peverada and William Croft]
Second International Workshop on Designing Meaning Representations, COLING 2020. -
Event structure representation: Between verbs and argument structure constructions
[Pavlina Kalm, Michael Regan, William Croft]
First International Workshop on Designing Meaning Representations, ACL 2019. -
CRAFT Shared Tasks 2019 Overview -- Integrated Structure, Semantics, and Coreference
[William Baumgartner, ..., Michael Regan and Lawrence Hunter]
5th Workshop on BioNLP Open Shared Tasks, EMNLP 2019. -
Tense and aspect semantics for sentential AMR
[Lucia Donatelli, Michael Regan, William Croft, and Nathan Schneider]
SCiL, 2019. -
A dataset for resolving referring expressions in spoken dialogue via contextual query rewrites (CQR).
[Michael Regan, Pushpendre Rastogi, Arpit Gupta, and Lambert Mathias]
arXiv, 2019.
Paper Data -
AMR beyond the sentence: the multi-sentence AMR corpus.
[Tim O'Gorman, Michael Regan, Kira Griffitt, Ulf Hermjakob, Kevin Knight, and Martha Palmer]
COLING, 2018.
Semantic typology
-
Cross-linguistic semantic annotation: Reconciling the language-specific and the universal.
[Jens Van Gysel, Meagan Vigus, Pavlina Kalm, Sook-kyung Lee, Michael Regan, and William Croft]
First International Workshop on Designing Meaning Representations, ACL 2019. -
Linguistic typology meets Universal Dependencies.
[William Croft, Dawn Nordquist, Katherine Looney, and Michael Regan]
TLT 2017: 15th International Workshop on Treebanks and Linguistic Theories.
- Reviewer: COLING 2020, 2022, 2024; NAACL 2021, 2022; GURT 2023; ACL 2023; *SEM 2023, 2024
- Program committee: SigTyp 2020, 2021, 2023