ArgMed-Agents: Explainable Clinical Decision Reasoning with LLM Disscusion via Argumentation Schemes

TL;DR

ArgMed-Agents addresses two major barriers to LLM-driven clinical reasoning: inadequate deep reasoning and opaque decision processes. It introduces a multi-agent workflow where Generators propose arguments, Verifiers challenge them with critical questions, and a Reasoner uses a symbolic solver to identify a coherent subset of arguments, forming a formal directed graph for decision support. The approach leverages ASCD within a non-monotonic reasoning framework to simulate clinical discussions and produce explanations, achieving higher accuracy than direct generation and Chain-of-Thought prompts, while boosting explainability toward knowledge-based CDSS benchmarks. Experimental results on MedQA and PubMedQA demonstrate improved performance and interpretability, and the work provides a conjecture about theoretical guarantees and boundaries of LLM capabilities in clinical reasoning, offering a pathway to safer, more auditable AI in healthcare.

Abstract

There are two main barriers to using large language models (LLMs) in clinical reasoning. Firstly, while LLMs exhibit significant promise in Natural Language Processing (NLP) tasks, their performance in complex reasoning and planning falls short of expectations. Secondly, LLMs use uninterpretable methods to make clinical decisions that are fundamentally different from the clinician's cognitive processes. This leads to user distrust. In this paper, we present a multi-agent framework called ArgMed-Agents, which aims to enable LLM-based agents to make explainable clinical decision reasoning through interaction. ArgMed-Agents performs self-argumentation iterations via Argumentation Scheme for Clinical Discussion (a reasoning mechanism for modeling cognitive processes in clinical reasoning), and then constructs the argumentation process as a directed graph representing conflicting relationships. Ultimately, use symbolic solver to identify a series of rational and coherent arguments to support decision. We construct a formal model of ArgMed-Agents and present conjectures for theoretical guarantees. ArgMed-Agents enables LLMs to mimic the process of clinical argumentative reasoning by generating explanations of reasoning in a self-directed manner. The setup experiments show that ArgMed-Agents not only improves accuracy in complex clinical decision reasoning problems compared to other prompt methods, but more importantly, it provides users with decision explanations that increase their confidence.

Figures (3)

• Figure 1: An example of Argumentation Scheme for Decision Making
• Figure 2: An example from the MedQA USMLE dataset, with the entire process of ArgMed-Agents reasoning about the clinical problem. Notably, the letters in the argumentation framework correspond to the serial numbers of the four generators on the right, representing the premises and conclusion generated by that generator. In the argumentation framework, the red nodes ($A$ and $C$) represent arguments in support of the decision and the yellow nodes ($B$ and $D$) represent arguments in support of the beliefs.
• Figure :

Theorems & Definitions (5)

• Definition 1: ArgMed-Agents Interaction
• Definition 2: Argumentation in ArgMed-Agents
• Definition 3
• Definition 4
• Example 1: From MedQA