AI conference NeurIPS extends paper submission deadline amid BLM protests

The deadline for submitting papers to AI mega-conference NeurIPS has been extended by 48 hours, to give people affected by the ongoing protests in the US more time to finish off their work. 

The new deadline for submissions to the conference, which starts December 5, is 1pm PDT on Friday, June 5.

In its announcement of the extension, the NeurIPS board said:

Today, NeurIPS grieves for its Black community members devastated by the cycle of police and vigilante violence.
Today, NeurIPS mourns for George Floyd, Breonna Taylor, Ahmaud Arbery, Regis Korchinski-Paquet, and thousands of Black people who have lost their lives to this violence.
Today, NeurIPS stands with its Black community to affirm that, today and every day, Black lives matter.

In a further show of support, researchers Erin Grant and Nicholas Le Rouz have created a list of AI mentors that black people working on papers can contact for advice. For every person getting help, both Le Roux and Google AI chief Jeff Dean have pledged to donate $1,000 to Black in AI, a group that aims to increase the presence of black people in artificial intelligence.

[Read: 7 ways to keep your personal information safe during protests]

An animation of the spreadsheet’s growth shows how quickly the community responded:

Would you look at that. pic.twitter.com/averQZF0d6

— Nicolas Le Roux (@le_roux_nicolas) June 2, 2020

Diversifying NeurIPS

The deadline extension is further evidence that NeurIPS wants to be a more inclusive event.

In recent years, the organizers have appointed a diversity and inclusion committee, diversified the speaker lineup, and hosted a growing range of diversity-related events, in response to widespread reports of discrimination at the conference.

They’ve also changed the event’s name from NIPS (Neural Information Processing System), after years of complaints that the acronym was both fostering sexist jokes and a slur for Japanese people.

While there’s still a huge diversity problem in AI, the changes taking place at NeurIPS show some progress is being made.