Scholar
Tianyi Lorena Yan
Google Scholar ID: uW7Ja9gAAAAJ
Columbia University
Citations & Impact
Citations
106
H-index
4
i10-index
3
Publications
5
Co-authors
0
Publications
Resume (English only)
Academic Achievements
- - Paper 'Promote, Suppress, Iterate: How Language Models Answer One-to-Many Factual Queries' accepted at EMNLP 2025
- - Participated in developing MuirBench, a comprehensive benchmark for robust multi-image understanding, accepted at ICLR 2025
- - Paper 'Contrastive Instruction Tuning' accepted at ACL 2024
- - Paper 'Monotonic Paraphrasing Improves Generalization of Language Model Prompting' accepted at EMNLP 2024
- - Paper 'Robust Natural Language Understanding with Residual Attention Debiasing' accepted at ACL 2023
Research Experience
- - May 2023 – August 2023: Diffusion-based Molecule Generation Research Intern at Tsinghua AIR
- - June 2022 – August 2022: M365 Deployment Software Engineer Intern at Microsoft
- - January 2022 – May 2024: Teaching Assistant at USC Viterbi School of Engineering, taught CSCI467 (Intro to Machine Learning) and CSCI270 (Intro to Algorithms and Theory of Computations)
- - February 2021 – May 2023: Leader of USC CSSA Career Development Mentorship Program
Education
- - PhD student at Columbia University, 2025-present, Advisor: Prof. John Hewitt
- - Bachelor's degree from University of Southern California (USC), Advisors: Prof. Robin Jia and Prof. Muhao Chen, and PhD mentor: Prof. Fei Wang
- - Collaborated with Prof. Hao Zhou at AIR Tsinghua
Background
- - Research Interests: Understanding the behaviors of language models and making them more controllable, interpretable, and efficient
- - Undergraduate at University of Southern California (USC), worked with Prof. Robin Jia and Prof. Muhao Chen
- - PhD advisor: Prof. John Hewitt at Columbia University
Miscellany
- - In her free time, she enjoys cooking, playing the flute (occasionally piccolo; played for 8 years!), and listening to soundtracks from movies and games
Co-authors