DS@GT at Touché: Large Language Models for Retrieval-Augmented Debate

Anthony Miyaguchi, Conor Johnston, Aaryan Potdar
7/12/2025
cs.IRcs.CL

Official code: dsgt-arc/touche-2025-rad (0 stars, MIT, last commit 2025-09-09).

Abstract

Large Language Models (LLMs) demonstrate strong conversational abilities. In this Working Paper, we study them in the context of debating in two ways: their ability to perform in a structured debate along with a dataset of arguments to use and their ability to evaluate utterances throughout the debate. We deploy six leading publicly available models from three providers for the Retrieval-Augmented Debate and Evaluation. The evaluation is performed by measuring four key metrics: Quality, Quantity, Manner, and Relation. Throughout this task, we found that although LLMs perform well in debates when given related arguments, they tend to be verbose in responses yet consistent in evaluation. The accompanying source code for this paper is located at https://github.com/dsgt-arc/touche-2025-rad.

View on arXivView PDF

Code Implementations(1)

dsgt-arc/touche-2025-radOfficial100%
MIT
01Jinja, ShellJan 17, 20251 years agoMIT

Similar Papers

An Empirical Analysis on Large Language Models in Debate Evaluation

89% match1 repo32024

DebateBench: A Challenging Long Context Reasoning Benchmark For Large Language Models

88% match2025

Evaluating the Performance of Large Language Models via Debates

87% match2024

Beyond Single Models: Enhancing LLM Detection of Ambiguity in Requests through Debate

86% match2025

Multi-LLM Debate: Framework, Principals, and Interventions

85% match2026

Large Language Models in Argument Mining: A Survey

85% match2025

Cite this paper

@article{miyaguchi2025dsgt,
  title  = {DS@GT at Touché: Large Language Models for Retrieval-Augmented Debate},
  author = {Anthony Miyaguchi and Conor Johnston and Aaryan Potdar},
  year   = {2025},
  eprint = {2507.09090},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2507.09090v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2507.09090.svg)](https://codeofpaper.com/paper/2507.09090)

Discussion