Towards More Faithful Natural Language Explanation Using Multi-Level Contrastive Learning in VQA

1 code implementation • 21 Dec 2023 • Chengen Lai, Shengli Song, Shiqi Meng, Jingyang Li, Sitong Yan, GuangNeng Hu

To address the above issues, we propose a novel self-supervised \textbf{M}ulti-level \textbf{C}ontrastive \textbf{L}earning based natural language \textbf{E}xplanation model (MCLE) for VQA with semantic-level, image-level, and instance-level factual and counterfactual samples.

Contrastive Learning counterfactual +3