DIRCR: Dual-Inference Rule-Contrastive Reasoning for Solving RAVENs
For abstract visual reasoning tasks, DIRCR addresses the bottleneck of incomplete rule capture by combining dual-inference paths and contrastive learning.
DIRCR integrates local row-wise and global holistic reasoning via a gated attention mechanism and uses rule-contrastive learning to improve feature separability, achieving significant gains on RAVEN datasets.
Abstract visual reasoning remains challenging as existing methods often prioritize either global context or local row-wise relations, failing to integrate both, and lack intermediate feature constraints, leading to incomplete rule capture and entangled representations. To address these issues, we propose the Dual-Inference Rule-Contrastive Reasoning (DIRCR) model. Its core component, the Dual-Inference Reasoning Module, combines a local path for row-wise analogical reasoning and a global path for holistic inference, integrated via a gated attention mechanism. Additionally, a Rule-Contrastive Learning Module introduces pseudo-labels to construct positive and negative rule samples, applying contrastive learning to enhance feature separability and promote abstract, transferable rule learning. Experimental results on three RAVEN datasets demonstrate that DIRCR significantly enhances reasoning robustness and generalization. Codes are available at https://github.com/csZack-Zhang/DIRCR.