$ whoami
Yilin Gong
PhD Student · Information Science · Luddy School, Indiana University Bloomington
advisor: Prof. Siqi Wu · previously: MS Data Science, University of Chicago
$ cat about.md
I study crowdsourced fact-checking and the governance of online information.
Current focus: a large-scale audit of X's Community Notes · contributor labor, topic coverage, and the effectiveness-fairness paradox.
Tools: Python · R · topic modeling · network analysis · mixed-effects models · LLMs.
Venues: ICWSM · AAAI · IC2S2.
$ ls papers/ --long
[2026]
ShareChatInsights: An Automated Pipeline for Analyzing Large-Scale Human–LLM Conversations
An AI agent turns open-ended questions about ShareChat, a corpus of human–LLM conversations across five commercial assistants, into fully specified studies (identify, scope, model, analyze) and exports each one as a runnable script. Evaluated on 46 requests drawn from prior literature.
[2026]
Characterizing AI Fact-Checkers and Their Contributions on Community Notes
20 AI systems wrote ~14% of Community Notes (rising to 45% recently), responding within minutes and filling 74.4% of gaps in human coverage — proportionally more helpful than lay contributors but still trailing experts.
[2026]
The Effects of Request Alerts on the Diversity and Visibility of Community Notes
Request Alerts redirect existing contributor labor rather than generating new effort, with a quality-distance trade-off when writers venture outside their topic expertise.
$ contact
yg24@iu.edu
· [cv]
· [scholar]
· [github]
· [linkedin]