Researchers Reduce Polarizing Posts' Visibility And Impact
Researchers developed and released an open-source web tool that reranked consenting participants' X (formerly Twitter) feeds for 10 days before the 2024 U.S. presidential election to deprioritize posts identified as polarizing by a large language model. The intervention lowered users' exposure to anti-democratic and partisan posts and measurably improved feelings toward opposing-party members while reducing negative emotions, with effects consistent across political affiliations.
Key Points
- 1Developed and deployed an open-source tool reranking X feeds to deprioritize polarizing posts during a 10-day trial.
- 2Found reduced exposure improved feelings toward opposing party and lowered negative emotions across political affiliations.
- 3Suggests platforms can use LLM-based classifiers and reranking to reduce partisan animosity without removing content.
Scoring Rationale
Field-relevant experimental evidence and open-source tools; limited by short 10-day trial and single-platform deployment scope.
Sources
Public references used for this report.
Practice with real Logistics & Shipping data
90 SQL & Python problems · 15 industry datasets
250 free problems · No credit card
See all Logistics & Shipping problems
