tjunlp-lab / commonmt

The Box is in the Pen: Evaluating Commonsense Reasoning in Neural Machine Translation

---

This is the official repo for the Findings of EMNLP 2020 paper: The Box is in the Pen: Evaluating Commonsense Reasoning in Neural Machine Translation .

This repository contains three types of test suite:

• lexical ambiguity

• contextless syntactic ambiguity

• contextual syntactic ambiguity

Data License

Attribution-ShareAlike 4.0 International (CC BY-SA 4.0) license. (License URL: https://creativecommons.org/licenses/by-sa/4.0/)

commonmt's Issues

Dataset used in table 2

Hi!  I noticed that in Table 2 of your paper, you used Commonsense Knowledge data set to make statistics with the proposed test set. I wonder if you could share this Commonsense Knowledge data set?  Thanks!