Katherine Tian

I'm a recent graduate from Harvard with a B.A. in CS & Statistics and M.S. in CS. I currently work on language model calibration and factuality at Stanford's Iris Lab, advised by Professor Chelsea Finn. Previously, at Harvard, I worked on medical vision-language models with Professor Pranav Rajpurkar, and algorithms for personal healthcare with Dr. Raaz Dwivedi (now Professor at Cornell Tech) and Professor Susan Murphy. I also interned at Google Brain where I worked on TensorFlow.

On campus, I was co-head TA for our machine learning class (CS 181) and TF'ed AI Research Experiences (CS 197). I also served as Career & Academics Director for Women in CS at Harvard and President of MITxHarvard Women in AI.

Email  /  Google Scholar  /  Linkedin  /  Twitter

profile photo

Research

Fine-Tuning Language Models for Factuality
Katherine Tian*, Eric Mitchell*, Huaxiu Yao, Christopher D. Manning, Chelsea Finn
International Conference on Learning Representations (ICLR), 2024
[arXiv]
Just Ask for Calibration: Strategies for Eliciting Calibrated Confidence Scores from Language Models Fine-Tuned with Human Feedback
Katherine Tian*, Eric Mitchell*, Allan Zhou, Archit Sharma, Rafael Rafailov, Huaxiu Yao, Chelsea Finn, Christopher D. Manning
Conference on Empirical Methods in Natural Language Processing (EMNLP), 2023
[arXiv]
Multimodal Image-Text Matching Improves Retrieval-based Chest X-Ray Report Generation
Jaehwan Jeong*, Katherine Tian*, Andrew Li, Sina Hartung, Fardad Behzadi, Juan Calle, David Osayande, Michael Pohlen, Subathra Adithan, Pranav Rajpurkar
Medical Imaging with Deep Learning (MIDL), 2023
[arXiv]
Counterfactual inference for sequential experimental design
Raaz Dwivedi, Katherine Tian, Sabina Tomkins, Predrag Klasnja, Susan Murphy, Devavrat Shah
In submission, 2023
[arXiv]
Automated clear cell renal carcinoma grade classification with prognostic significance
Katherine Tian*, Christopher A Rubadue*, Douglas I Lin, Mitko Veta, Michael E Pyle, Humayun Irshad, Yujing J Heng
PloS one Vol. 14 (10), 2019
[link]
Technical Note on Transcription Factor Motif Discovery from Importance Scores (TF-MoDISco)
Avanti Shrikumar, Katherine Tian, Žiga Avsec, Anna Shcherbina, Abhimanyu Banerjee, Mahfuza Sharmin, Surag Nair, Anshul Kundaje
Preprint, 2018
[arXiv]

News

  • Apr 29, 2024, Invited talk at Amazon AWS Responsible AI on "Fine-Tuning Language Models for Factuality"
  • May 7-11, 2024, attending ICLR at Vienna Austria