✨News

• [Sep 2026] JudgmentBench: Comparing rubric and preference evaluation for quality assessment was accepted to the NeurIPS 2026 Evaluations & Datasets Track.
• [Aug 2026] Inducing task models from computer-use traces was accepted to EMNLP 2026.
• [Aug 2026] Presented Homogenization or epistemic pluralism: LLM-generated reviews in scientific peer review at the 2026 ASA Annual Meeting in New York.
• [Aug 2026] Selected as a Stanford HAI Data Science Scholar (2026–2028).
• [Apr 2026] Presented work on high school students’ AI usage patterns at AERA 2026 in Los Angeles.
• [Spring 2026] Received a Schmidt Sciences Proposal Development Award with Daniel A. McFarland and Tianyu Du.