Evaluating Uncertainty Quantification Methods in Argumentative Large Language Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhou, Kevin, Dejl, Adam, Freedman, Gabriel, Chen, Lihu, Rago, Antonio, Toni, Francesca
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913097788162048
author Zhou, Kevin
Dejl, Adam
Freedman, Gabriel
Chen, Lihu
Rago, Antonio
Toni, Francesca
author_facet Zhou, Kevin
Dejl, Adam
Freedman, Gabriel
Chen, Lihu
Rago, Antonio
Toni, Francesca
contents Research in uncertainty quantification (UQ) for large language models (LLMs) is increasingly important towards guaranteeing the reliability of this groundbreaking technology. We explore the integration of LLM UQ methods in argumentative LLMs (ArgLLMs), an explainable LLM framework for decision-making based on computational argumentation in which UQ plays a critical role. We conduct experiments to evaluate ArgLLMs' performance on claim verification tasks when using different LLM UQ methods, inherently performing an assessment of the UQ methods' effectiveness. Moreover, the experimental procedure itself is a novel way of evaluating the effectiveness of UQ methods, especially when intricate and potentially contentious statements are present. Our results demonstrate that, despite its simplicity, direct prompting is an effective UQ strategy in ArgLLMs, outperforming considerably more complex approaches.
format Preprint
id arxiv_https___arxiv_org_abs_2510_02339
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Evaluating Uncertainty Quantification Methods in Argumentative Large Language Models
Zhou, Kevin
Dejl, Adam
Freedman, Gabriel
Chen, Lihu
Rago, Antonio
Toni, Francesca
Computation and Language
Artificial Intelligence
Research in uncertainty quantification (UQ) for large language models (LLMs) is increasingly important towards guaranteeing the reliability of this groundbreaking technology. We explore the integration of LLM UQ methods in argumentative LLMs (ArgLLMs), an explainable LLM framework for decision-making based on computational argumentation in which UQ plays a critical role. We conduct experiments to evaluate ArgLLMs' performance on claim verification tasks when using different LLM UQ methods, inherently performing an assessment of the UQ methods' effectiveness. Moreover, the experimental procedure itself is a novel way of evaluating the effectiveness of UQ methods, especially when intricate and potentially contentious statements are present. Our results demonstrate that, despite its simplicity, direct prompting is an effective UQ strategy in ArgLLMs, outperforming considerably more complex approaches.
title Evaluating Uncertainty Quantification Methods in Argumentative Large Language Models
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2510.02339