Adversarially Regularising Neural NLI Models to Integrate Logical Background Knowledge

Open in new window