Extracting a Prototypical Argumentative Pattern in Financial Q&As

Giulia D'Agostino; Michiel van der Meer; Chris Reed

Extracting a Prototypical Argumentative Pattern in Financial Q&As

Giulia D'Agostino
, Michiel van der Meer
, Chris Reed

Computing

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Argumentative patterns are recurrent strategies adopted to pursue a definite communicative goal in a discussion. For instance, in Q&A exchanges during financial conference calls, a pattern called Request of Confirmation of Inference (ROCOI) helps streamline conversations by requesting explicit verification of inferences drawn from a statement. Our work presents two ROCOI extraction approaches from interrogative units: sequence labeling and text-to-text generation. We experiment with multiple models for each task formulation to explore which models can effectively and robustly perform pattern extraction. Results indicate that machine-based ROCOI extraction is an achievable task, though variation among metrics that are designed for different evaluation dimensions makes obtaining a clear picture difficult. We find that overall, ROCOI extraction is performed best via sequence labeling, though with ample room for improvement. We encourage future work to extend the study to new argumentative patterns.

Original language	English
Title of host publication	Proceedings of the 2025 CLASP Conference on Language models and RePresentations (LARP
Editors	Nikolai Ilinykh, Mattias Appelgren, Erik Lagerstedt
Place of Publication	Pennsylvania, USA
Publisher	Association for Computational Linguistics
Pages	51–64
Number of pages	14
ISBN (Print)	979-8-89176-249-7
Publication status	Published - Sept 2025
Event	2025 CLASP conference on Language Models and RePresentations - University in Gothenburg, Gothenburg, Sweden Duration: 8 Sept 2025 → 9 Sept 2025 https://gu-clasp.github.io/LARP/

Conference

Conference	2025 CLASP conference on Language Models and RePresentations
Abbreviated title	LARP 2025
Country/Territory	Sweden
City	Gothenburg
Period	8/09/25 → 9/09/25
Internet address	https://gu-clasp.github.io/LARP/

Access to Document

Final Published VersionFinal published version, 257 KBLicence: CC BY

https://aclanthology.org/2025.clasp-main.5/Licence: CC BY

Cite this

@inproceedings{7fbdeeb11e8f4daaab76dd7da00644ce,

title = "Extracting a Prototypical Argumentative Pattern in Financial Q\&As",

abstract = "Argumentative patterns are recurrent strategies adopted to pursue a definite communicative goal in a discussion. For instance, in Q\&A exchanges during financial conference calls, a pattern called Request of Confirmation of Inference (ROCOI) helps streamline conversations by requesting explicit verification of inferences drawn from a statement. Our work presents two ROCOI extraction approaches from interrogative units: sequence labeling and text-to-text generation. We experiment with multiple models for each task formulation to explore which models can effectively and robustly perform pattern extraction. Results indicate that machine-based ROCOI extraction is an achievable task, though variation among metrics that are designed for different evaluation dimensions makes obtaining a clear picture difficult. We find that overall, ROCOI extraction is performed best via sequence labeling, though with ample room for improvement. We encourage future work to extend the study to new argumentative patterns.",

author = "Giulia D'Agostino and \{van der Meer\}, Michiel and Chris Reed",

note = "{\textcopyright} 2025 Association for Computational Linguistics; 2025 CLASP conference on Language Models and RePresentations, LARP 2025 ; Conference date: 08-09-2025 Through 09-09-2025",

year = "2025",

month = sep,

language = "English",

isbn = "979-8-89176-249-7",

pages = "51–64",

editor = "Ilinykh, \{Nikolai \} and Appelgren, \{Mattias \} and Lagerstedt, \{Erik \}",

booktitle = "Proceedings of the 2025 CLASP Conference on Language models and RePresentations (LARP",

publisher = "Association for Computational Linguistics",

url = "https://gu-clasp.github.io/LARP/",

}

D'Agostino, G, van der Meer, M & Reed, C 2025, Extracting a Prototypical Argumentative Pattern in Financial Q&As. in N Ilinykh, M Appelgren & E Lagerstedt (eds), Proceedings of the 2025 CLASP Conference on Language models and RePresentations (LARP. Association for Computational Linguistics, Pennsylvania, USA, pp. 51–64, 2025 CLASP conference on Language Models and RePresentations, Gothenburg, Sweden, 8/09/25. <https://aclanthology.org/2025.clasp-main.5/>

Extracting a Prototypical Argumentative Pattern in Financial Q&As. / D'Agostino, Giulia; van der Meer, Michiel; Reed, Chris.
Proceedings of the 2025 CLASP Conference on Language models and RePresentations (LARP. ed. / Nikolai Ilinykh; Mattias Appelgren; Erik Lagerstedt. Pennsylvania, USA: Association for Computational Linguistics, 2025. p. 51–64.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Extracting a Prototypical Argumentative Pattern in Financial Q&As

AU - D'Agostino, Giulia

AU - van der Meer, Michiel

AU - Reed, Chris

PY - 2025/9

Y1 - 2025/9

N2 - Argumentative patterns are recurrent strategies adopted to pursue a definite communicative goal in a discussion. For instance, in Q&A exchanges during financial conference calls, a pattern called Request of Confirmation of Inference (ROCOI) helps streamline conversations by requesting explicit verification of inferences drawn from a statement. Our work presents two ROCOI extraction approaches from interrogative units: sequence labeling and text-to-text generation. We experiment with multiple models for each task formulation to explore which models can effectively and robustly perform pattern extraction. Results indicate that machine-based ROCOI extraction is an achievable task, though variation among metrics that are designed for different evaluation dimensions makes obtaining a clear picture difficult. We find that overall, ROCOI extraction is performed best via sequence labeling, though with ample room for improvement. We encourage future work to extend the study to new argumentative patterns.

AB - Argumentative patterns are recurrent strategies adopted to pursue a definite communicative goal in a discussion. For instance, in Q&A exchanges during financial conference calls, a pattern called Request of Confirmation of Inference (ROCOI) helps streamline conversations by requesting explicit verification of inferences drawn from a statement. Our work presents two ROCOI extraction approaches from interrogative units: sequence labeling and text-to-text generation. We experiment with multiple models for each task formulation to explore which models can effectively and robustly perform pattern extraction. Results indicate that machine-based ROCOI extraction is an achievable task, though variation among metrics that are designed for different evaluation dimensions makes obtaining a clear picture difficult. We find that overall, ROCOI extraction is performed best via sequence labeling, though with ample room for improvement. We encourage future work to extend the study to new argumentative patterns.

UR - https://aclanthology.org/2025.clasp-main/

M3 - Conference contribution

SN - 979-8-89176-249-7

SP - 51

EP - 64

BT - Proceedings of the 2025 CLASP Conference on Language models and RePresentations (LARP

A2 - Ilinykh, Nikolai

A2 - Appelgren, Mattias

A2 - Lagerstedt, Erik

PB - Association for Computational Linguistics

CY - Pennsylvania, USA

T2 - 2025 CLASP conference on Language Models and RePresentations

Y2 - 8 September 2025 through 9 September 2025

ER -

Extracting a Prototypical Argumentative Pattern in Financial Q&As

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this