Language Models Are Poor Learners of Directional Inference.

Tianyi Li, Mohammad Javad Hosseini, Sabine Weber, Mark Steedman

Browse the full EMNLP paper archive.