Research
Publications & preprints.
Work on misinformation, digital platforms, and the social dimensions of AI.
Peer-reviewed publications
Where Do LLMs Go Wrong? Diagnosing Automated Peer Review via Aspect-Guided Multi-Level Perturbation
Diagnosing the robustness and biases of LLM reviewers through controlled changes to papers, reviews, and rebuttals.
Preprints
Analyzing Cognitive Differences Among Large Language Models through the Lens of Social Worldview
Examining models’ orientations toward hierarchy, equality, individuality, and fate—and their responses to social cues.
Working papers
A Human-LLM Framework for Analyzing Amplified Crisis Comments on YouTube
Combining human interpretation, language models, and network analysis to understand how crisis conversations are amplified on video platforms.