Compact Prompting in Instruction-tuned LLMs for Joint Argumentative Component Detection

TL;DR

This paper introduces a novel method for Argumentative Component Detection using instruction-tuned LLMs with compact prompts, framing it as a generative task to improve performance over existing approaches.

Contribution

It is the first to model ACD as a generative task with instruction-tuned LLMs, moving beyond traditional classification and segmentation methods.

Findings

Outperforms state-of-the-art systems on standard benchmarks

Demonstrates the effectiveness of instruction tuning for complex argument mining tasks

Shows that ACD can be effectively modeled as a language generation problem

Abstract

Argumentative component detection (ACD) is a core subtask of Argument(ation) Mining (AM) and one of its most challenging aspects, as it requires jointly delimiting argumentative spans and classifying them into components such as claims and premises. While research on this subtask remains relatively limited compared to other AM tasks, most existing approaches formulate it as a simplified sequence labeling problem, component classification, or a pipeline of component segmentation followed by classification. In this paper, we propose a novel approach based on instruction-tuned Large Language Models (LLMs) using compact instruction-based prompts, and reframe ACD as a language generation task, enabling arguments to be identified directly from plain text without relying on pre-segmented components. Experiments on standard benchmarks show that our approach achieves higher performance compared to state-of-the-art systems. To the best of our knowledge, this is one of the first attempts to fully model ACD as a generative task, highlighting the potential of instruction tuning for complex AM problems.