Can RNNs trained on harder subject-verb agreement instances still perform well on easier ones? Hritik Bansal author Gantavya Bhatt author Sumeet Agarwal author 2021-02 text Proceedings of the Society for Computation in Linguistics 2021 Allyson Ettinger editor Ellie Pavlick editor Brandon Prickett editor Association for Computational Linguistics Online conference publication bansal-etal-2021-rnns https://aclanthology.org/2021.scil-1.37/ 2021-02 366 373