Your submission is now in Draft mode.

Once it's ready, please submit your draft for review by our team of Community Moderators. Thank you!

Submit Essay

Once you submit your essay, you can no longer edit it.

Pending

This content now needs to be approved by community moderators.

Submitted

This essay was submitted and is waiting for review.

What will be the best score in the 2019/2020 Winograd Schema AI challenge

Question

The "Winograd Schema Challenge" was devised by Hector Levesque, to test the ability of a machine intelligences system to understand the meaning of various sentences in a way that goes beyond pattern matching of syntactics. A question might be, for example:

Babar wonders how he can get new clothing. Luckily, a very rich old man who has always been fond of little elephants understands right away that he is longing for a fine suit. As he likes to make people happy, he gives him his wallet.

Who is longing for a fine suit? Who likes to make people happy? Who gives a wallet? Who is a wallet given to?

These questions cannot be answered effectively by the sort of pattern matching that underlies much machine transcription, translation, etc., but humans can do them fairly readily (with a typical human scoring 90%+ on this sort of test).

Machine systems are tested against this schema in annual competitions sponsored by Nuance and organized by coomonsensereasoning.org. In the latest competition, the best competitor scores about 60%. (Because many of the questions are binary, this is not a terribly impressive score; the full 2016 question list is here.)

The 2018 Challenge did not take place due to several late cancellations. The timing of the next challenge is unclear, but seems likely to be either 2019 or 2020.

What score, in percentage correct, will the top entrant in the 2019 or 2020 (whichever is the next after 2016) Winograd challenge score?

There are two rounds, and this applies to the first round only. Resolves as ambiguous if the test is not held by end of 2020, or if there are significant changes from the 2016 competition so that the scores are not directly comparable. Closes retroactively to noon one day before the listed date of the competition.

Make a Prediction

Prediction

Note: this question resolved before its original close time. All of your predictions came after the resolution, so you did not gain (or lose) any points for it.

Note: this question resolved before its original close time. You earned points up until the question resolution, but not afterwards.

Current points depend on your prediction, the community's prediction, and the result. Your total earned points are averaged over the lifetime of the question, so predict early to get as many points as possible! See the FAQ.