Adaptive Collaboration of Arena-Based Argumentative LLMs for Explainable and Contestable Legal Reasoning

TL;DR

This paper introduces ACAL, a neuro-symbolic framework that enhances legal reasoning in LLMs by enabling structured, contestable, and human-auditable arguments through multi-agent collaboration and argumentation frameworks.

Contribution

ACAL is the first framework to integrate adaptive multi-agent collaboration with an argumentation framework for explainable legal reasoning in LLMs.

Findings

ACAL outperforms baseline models on LegalBench.

It balances accuracy with transparency and contestability.

Empowers human users to audit and modify reasoning processes.

Abstract

Legal reasoning requires not only high accuracy but also the ability to justify decisions through verifiable and contestable arguments. However, existing Large Language Model (LLM) approaches, such as Chain-of-Thought (CoT) and Retrieval-Augmented Generation (RAG), often produce unstructured explanations that lack a formal mechanism for verification or user intervention. To address this limitation, we propose Adaptive Collaboration of Argumentative LLMs (ACAL), a neuro-symbolic framework that integrates adaptive multi-agent collaboration with an Arena-based Quantitative Bipolar Argumentation Framework (A-QBAF). ACAL dynamically deploys expert agent teams to construct arguments, employs a clash resolution mechanism to adjudicate conflicting claims, and utilizes uncertainty-aware escalation for borderline cases. Crucially, our framework supports a Human-in-the-Loop (HITL) contestability workflow, enabling users to directly audit and modify the underlying reasoning graph to influence the final judgment. Empirical evaluations on the LegalBench benchmark demonstrate that ACAL outperforms strong baselines across Gemini-2.5-Flash-Lite and Gemini-2.5-Flash architectures, effectively balancing efficient predictive performance with structured transparency and contestability. Our implementation is available at: https://github.com/loc110504/ACAL.