Diff-Explainer: Differentiable Convex Optimization for Explainable Multi-hop Inference

26723

Knowledge Triples