Explaining Arguments' Strength: Unveiling the Role of Attacks and Supports (Technical Report)
Xiang Yin, Potyka Nico, Francesca Toni
TL;DR
This work introduces Relation Attribution Explanations (RAEs) to explain the strength of topic arguments in Quantitative Bipolar AFs (QBAFs) under gradual semantics by adapting Shapley values to edge contributions. RAEs assign edge-level contributions to both attacks and supports, including indirect paths, and are equipped with a suite of properties (Shapley-based and argumentative) along with a probabilistic approximation algorithm that converges to the true values. Two case studies—Fraud Detection and Large Language Models (LLMs)—demonstrate RAEs’ practical utility, revealing nuanced, path-specific influences that go beyond traditional argument-level attributions. The approach provides a principled, interpretable framework for explaining QBAFs and suggests future work on joint Shapley analyses, edge-weighted QBAFs, and user-centered evaluation.
Abstract
Quantitatively explaining the strength of arguments under gradual semantics has recently received increasing attention. Specifically, several works in the literature provide quantitative explanations by computing the attribution scores of arguments. These works disregard the importance of attacks and supports, even though they play an essential role when explaining arguments' strength. In this paper, we propose a novel theory of Relation Attribution Explanations (RAEs), adapting Shapley values from game theory to offer fine-grained insights into the role of attacks and supports in quantitative bipolar argumentation towards obtaining the arguments' strength. We show that RAEs satisfy several desirable properties. We also propose a probabilistic algorithm to approximate RAEs efficiently. Finally, we show the application value of RAEs in fraud detection and large language models case studies.