1 code implementation • Joint Conference on Lexical and Computational Semantics 2021 • Ohad Rozen, Shmuel Amar, Vered Shwartz, Ido Dagan
Our approach facilitates learning generic inference patterns requiring relational knowledge (e. g. inferences related to hypernymy) during training, while injecting on-demand the relevant relational facts (e. g. pangolin is an animal) at test time.
no code implementations • NAACL 2021 • Ohad Rozen, David Carmel, Avihai Mejer, Vitaly Mirkis, Yftah Ziser
In this work, we propose a novel and complementary approach for predicting the answer for such questions, based on the answers for similar questions asked on similar products.
1 code implementation • CONLL 2019 • Ohad Rozen, Vered Shwartz, Roee Aharoni, Ido Dagan
Phenomenon-specific "adversarial" datasets have been recently designed to perform targeted stress-tests for particular inference types.
