Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

LLM-based Argument Mining meets Argumentation and Description Logics: a Unified Framework for Reasoning about Debates

Gianvincenzo Alfano, Sergio Greco, Lucio La Cava, Stefano Francesco Monea, Irina Trubitsyna

TL;DR

This work proposes a framework that integrates learning-based argument mining with quantitative reasoning and ontology-based querying, and provides a transparent, explainable, and formally grounded method for analyzing debates, overcoming purely statistical LLM-based analyses.

Abstract

Large Language Models (LLMs) achieve strong performance in analyzing and generating text, yet they struggle with explicit, transparent, and verifiable reasoning over complex texts such as those containing debates. In particular, they lack structured representations that capture how arguments support or attack each other and how their relative strengths determine overall acceptability. We encompass these limitations by proposing a framework that integrates learning-based argument mining with quantitative reasoning and ontology-based querying. Starting from a raw debate text, the framework extracts a fuzzy argumentative knowledge base, where arguments are explicitly represented as entities, linked by attack and support relations, and annotated with initial fuzzy strengths reflecting plausibility w.r.t. the debate's context. Quantitative argumentation semantics are then applied to compute final argument strengths by propagating the effects of supports and attacks. These results are then embedded into a fuzzy description logic setting, enabling expressive query answering through efficient rewriting techniques. The proposed approach provides a transparent, explainable, and formally grounded method for analyzing debates, overcoming purely statistical LLM-based analyses.

LLM-based Argument Mining meets Argumentation and Description Logics: a Unified Framework for Reasoning about Debates

TL;DR

This work proposes a framework that integrates learning-based argument mining with quantitative reasoning and ontology-based querying, and provides a transparent, explainable, and formally grounded method for analyzing debates, overcoming purely statistical LLM-based analyses.

Abstract

Large Language Models (LLMs) achieve strong performance in analyzing and generating text, yet they struggle with explicit, transparent, and verifiable reasoning over complex texts such as those containing debates. In particular, they lack structured representations that capture how arguments support or attack each other and how their relative strengths determine overall acceptability. We encompass these limitations by proposing a framework that integrates learning-based argument mining with quantitative reasoning and ontology-based querying. Starting from a raw debate text, the framework extracts a fuzzy argumentative knowledge base, where arguments are explicitly represented as entities, linked by attack and support relations, and annotated with initial fuzzy strengths reflecting plausibility w.r.t. the debate's context. Quantitative argumentation semantics are then applied to compute final argument strengths by propagating the effects of supports and attacks. These results are then embedded into a fuzzy description logic setting, enabling expressive query answering through efficient rewriting techniques. The proposed approach provides a transparent, explainable, and formally grounded method for analyzing debates, overcoming purely statistical LLM-based analyses.
Paper Structure (15 sections, 4 theorems, 9 equations, 3 figures, 1 table)

This paper contains 15 sections, 4 theorems, 9 equations, 3 figures, 1 table.

Table of Contents

  1. Introduction
  2. Preliminaries
  3. Quantitative Bipolar AF
  4. Description Logics
  5. Argument Mining and LLMs
  6. Reasoning on Textual Dialogues
  7. Extracting FABox through LLMs
  8. Argument Tagging
  9. Entity Extractor
  10. QBAF Extractor
  11. FABox Builder
  12. Validation
  13. Reasoning in FAKB
  14. Related Work
  15. Conclusion and Future Work

Key Result

Proposition 1

For any consistent FABox $\mathcal{A}$, $\Lambda\xspace_{\!_\mathcal{A}}=\langle\hbox{$\tt Arg$}^{1},\hbox{$\tt att$}^{1},\hbox{$\tt sup$}^{1}, \hbox{$\tt Arg$}^{2}, \hbox{$\tt att$}^{2}\cup\hbox{$\tt sup$}^{2}\rangle$ forms a QBAF.

Figures (3)

  • Figure 1: FAKB workflow: from raw unstructured texts to FAKB generation and querying.
  • Figure 2: An example of updated FAbox, visualized in the form of a knowledge graph, and extracted from an extended version of debate $D$ of Example \ref{['ex:intro0']}. Nodes are color-coded by type: blue for entities, red for argument identifiers, and yellow for concepts. Similarly, edges represent specific relations: green/red denotes support/attacks, while dark blue, light blue, and yellow represent authorship, role instances, and concept instances, respectively. Upper/lower numbers within each argument denote its initial/updated strength. For optimal visualization, the arguments' text has been omitted, yet they can be found in Section \ref{['sec:validation']}. A full-screen visualization also including additional entities and full texts is provided in the Appendix.
  • Figure 3: An example of updated FAbox, visualized in the form of a knowledge graph, and extracted from an extended version of debate $D$ of Example \ref{['ex:intro0']}. Nodes are color-coded by type: violet for text of arguments, blue for entities, red for argument identifiers, and yellow for concepts. Similarly, edges represent specific relations: green/red denotes support/attacks, while dark blue, light blue, and yellow represent authorship, role instances, and concept instances, respectively. Upper/lower numbers within each argument denote its initial/updated strength.

Theorems & Definitions (15)

  • Example 1
  • Example 2
  • Example 3
  • Example 4
  • Definition 1
  • Definition 2
  • Proposition 1
  • Example 5
  • Definition 3
  • Example 6
  • ...and 5 more