HackerPapers

arXiv:2210.15723cs.SI

Birdwatch: Crowd Wisdom and Bridging Algorithms can Inform Understanding and Reduce the Spread of Misinformation

Stefan Wojcik, Sophie Hilgard, Nick Judd, Delia Mocanu, Stephen Ragain, M. B. Fallin Hunzaker, Keith Coleman, Jay Baxter

Abstract

We present an approach for selecting objectively informative and subjectively helpful annotations to social media posts. We draw on data from on an online environment where contributors annotate misinformation and simultaneously rate the contributions of others. Our algorithm uses a matrix-factorization (MF) based approach to identify annotations that appeal broadly across heterogeneous user groups - sometimes referred to as "bridging-based ranking." We pair these data with a survey experiment in which individuals are randomly assigned to see annotations to posts. We find that annotations selected by the algorithm improve key indicators compared with overall average and crowd-generated baselines. Further, when deployed on Twitter, people who saw annotations selected through this bridging-based approach were significantly less likely to reshare social media posts than those who did not see the annotations.

arXiv comments
Journal reference
DOI
Subjects
cs.SISocial and Information Networks

On Hacker News

Linked by 4 identifiable commenters in 6 comments across 4 threads, from Nov 2022 to Jan 2025. Found in 2 HN submissions from Oct 2022 to Jan 2024. Open submission discussions.

Comments linking this paper, per year.

Discussed in comments4 threads

  1. Ending our third party fact-checking program and moving to Community Notes model 3 comments

    • Here's the Community Notes whitepaper [1], for how it all works. Previous discussion [2]. [1] Birdwatch: Crowd Wisdom and Bridging Algorithms can Inform Understanding and Reduce the Spread of Misinformation, https://arxiv.org/abs/2210.15723 [2]…

      nomelView on HNParent context

    • > Community notes are just opinions of random people on internet It's much more complicated than that. Here's the white paper: https://arxiv.org/abs/2210.15723

      nomelView on HNParent context

    • A way to quantify this doesn't immediately come to my mind. Maybe reasonable metrics would be: 1. What % misleading/false posts are flagged 2. What % of those flagged are given meaningful context/corrections that are accurate. It seems there's circular logic of first determining…

      nomelView on HNParent context

  2. Gin, Television, and Social Surplus (2008) 1 comment

    • Hmm sorry don't have anything. I recall hearing about some international research project that was used as input when creating the Twitter community notes but can't find it anymore. Interesting graphical tools exist nowadays to navigate the citation web and I guess one could try…

      GravitylossView on HNParent context

  3. The Scramble to Save Twitter’s Research from Elon Musk 1 comment

    • A list of recent papers by Twitter-affiliated authors would be helpful, but as far as I can tell arxiv.org doesn't support that type of search. Here is the mentioned "Birdwatch" paper: https://arxiv.org/abs/2210.15723

      webmavenView on HNParent context

  4. White House deletes tweet after Twitter adds 'context' note 1 comment

    • Hi! Birdwatch ML Engineer here-- these context notes are from Birdwatch. They are written and rated by users, and the notes are only added as context to Tweets if they are rated highly enough by multiple raters who have tended to disagree in the past. The core algorithm is open…

      jaybaxterView on HNParent context

Submitted to Hacker News

Connected through shared HN threads, not inferred similarity or a recommendation.