Object-Centric Case-Based Reasoning via Argumentation
Gabriel de Olim Gaul, Adam Gould, Avinash Kori, Francesca Toni
TL;DR
This work presents Slot-Attention Argumentation for Case-Based Reasoning (SAA-CBR), a neuro-symbolic framework that combines object-centric Slot Attention with Abstract Argumentation for Case-Based Reasoning to enable explainable image classification. By integrating feature combination strategies, casebase reduction, counting-based partial orders, and multi-class extensions like One-Vs-Rest and Supported AA-CBR, the approach achieves competitive results on CLEVR-Hans datasets and demonstrates improved generalization over purely neural baselines. The paper provides a thorough experimental evaluation, ablation analyses, and discusses practical considerations for scaling and end-to-end learning, highlighting the potential of combining structured argumentation with neural representations for interpretable, robust AI. Overall, SAA-CBR advances explainable AI by grounding object-centric perceptions in symbolic reasoning, enabling transparent decision-making with tunable control over reasoning complexity and case usage.
Abstract
We introduce Slot Attention Argumentation for Case-Based Reasoning (SAA-CBR), a novel neuro-symbolic pipeline for image classification that integrates object-centric learning via a neural Slot Attention (SA) component with symbolic reasoning conducted by Abstract Argumentation for Case-Based Reasoning (AA-CBR). We explore novel integrations of AA-CBR with the neural component, including feature combination strategies, casebase reduction via representative samples, novel count-based partial orders, a One-Vs-Rest strategy for extending AA-CBR to multi-class classification, and an application of Supported AA-CBR, a bipolar variant of AA-CBR. We demonstrate that SAA-CBR is an effective classifier on the CLEVR-Hans datasets, showing competitive performance against baseline models.