Chenyan Jia
@jiachenyan.bsky.social
Assistant Professor @Northeastern. Previously @Stanford postdoc. @UTAustin @PKU1898 alum. Interested in human-centered AI, HCI, and misinformation.
Our new study in @science.org built an LLM-powered browser extension to rerank social media feeds without requiring platform cooperation. In a preregistered 10-day field experiment (N=1,256), we found that algorithmic ranking can both raise and lower levels of affective political polarization.
December 3, 2025 at 12:45 AM
Our new study in @science.org built an LLM-powered browser extension to rerank social media feeds without requiring platform cooperation. In a preregistered 10-day field experiment (N=1,256), we found that algorithmic ranking can both raise and lower levels of affective political polarization.