NLP / WIC Benchmark:

Up to 2900 words and 700 WiC sentences.

A problem with the WIC test is it isn't specific about in what context makes two sentences similar or not.

Before, I matched for different Intentions. The words before and after the highlighted word. Eg. 1: "for one person to do". 2: "each person is unique". Different intentions, but same word meaning.

Now it appears to mean Homophones. Any case where the word is the same it's a match, as above.

So it may help to have two lists. One Intentions. One Homophones.