About Jiayi (Raina) Xin
I’m a Ph.D. student in Computer and Information Science at University of Pennsylvania, advised by Prof. Qi Long and co-advised by Prof. Weijie Su. My research lies at the intersection of multimodal learning, large language models, and agentic AI, with a focus on statistical foundations, reliability, and biomedical applications.
News
- Oct 2026: Explore my new Research Blog for detailed walkthrough of our papers on multimodal learning, language models, and biomedical AI.
- Sep 2026: Two papers accepted to NeurIPS 2026 main track: Conformal Prediction with Paraphrase-Aware Scoring and CurveRL. See you in Atlanta, Dec. 8–13!
- Sep 2026: Presented UCS: Estimating Unseen Coverage for Improved In-Context Learning at the AWS ASSET Trustworthy AI Symposium at Penn.
Earlier news
- Aug 2026: Earned my M.S. in Computer & Information Science from the University of Pennsylvania with a 4.0/4.0 GPA, en route to my Ph.D.
- Aug 2026: Completed my research internship at Microsoft Research Health Futures, where I worked on AI for biology.
- Jul 2026: Presented our paper, “UCS: Estimating Unseen Coverage for Improved In-Context Learning,” at ACL 2026 in San Diego, CA.
- Jun 2026: Gave an oral presentation on “Conformal Prediction with Paraphrase-Aware Scoring for LLM Uncertainty Quantification” at the 2026 NISS Virtual New Researchers Conference.
- May 2026: Started my research internship at the MSR Health Futures Immunomics Group in Redmond, WA.
- Apr 2026: Was selected as a finalist of the IMC PhD Fellowship and attended the Research Connect event in Chicago, IL.
- Apr 2026: Our paper, UCS: Estimating Unseen Coverage for Improved In-Context Learning, was accepted to ACL 2026 Findings.
- Apr 2026: Presented I$^2$MoE as a poster at the Inaugural AASF AIX Summit in New York City, NY.
- Mar 2026: Presented KnowSum at ENAR 2026 in Indianapolis, IN.
- Jan 2026: Accepted a research internship offer from Microsoft Health Futures for May–Aug 2026 in Redmond, WA.
- Dec 2025: Excited to present Improved Therapeutic Antibody Reformatting through Multimodal Machine Learning as a workshop poster at NeurIPS 2025!
- Nov 2025: Excited to present I$^2$MoE as a Poster at the 10th Annual MidAtlantic Bioinformatics Conference!
- Sep 2025: Improved Therapeutic Antibody Reformatting through Multimodal Machine Learning accepted into NeurIPS AI4Science and FM4LifeSciences workshops!
- Sep 2025: Excited to present I$^2$MoE at Penn Engineering AI Mixer and Penn DBEI Big Data conference!
- Aug 2025: Language Mixing on Bilingual LLM Reasoning accepted as an Oral Presentation at EMNLP 2025!
- Jun 2025: Estimate LLM Unseen Knowledge with KnowSum is now on arXiv! KnowSum estimates how much unseen knowledge today's large language models know.
- May 2025: Started my internship as a Machine Learning Intern at BigHat Biosciences! Excited to apply my ML knowledge in solving challenges in antibody therapies.
- May 2025: Interpretable Multimodal Interaction-aware Mixture-of-Experts (I$^2$MoE) got accepted into ICML 2025! I$^2$MoE achieves superior multimodal fusion performance and reveals how multimodal interactions influence predictions.
- Feb 2025: Missing Modality Imputation with MoE-Retriever was awarded as the Best Paper at AAAI GenAI4Health Workshop 2025! MoE-Retriever fills in the gaps when some modalities, e.g., MRI scans and clinical notes, are missing.
Selected publications
I2MoE: Interpretable Multimodal Interaction-Aware Mixture-of-Experts
ICML 2025
Conformal Prediction with Paraphrase-Aware Scoring for LLM Uncertainty Quantification
NeurIPS 2026
UCS: Estimating Unseen Coverage for Improved In-Context Learning
ACL 2026 Findings
Academic Journey
- M.S. in Computer & Information Science, University of Pennsylvania
- GPA: 4.0/4.0; earned en route to the Ph.D., Aug 2024 – Aug 2026.
- B.A.Sc. in Applied AI, The University of Hong Kong
- Capstone with Prof. Lequan Yu & mentor Fuying Wang.
- Exchange & Research at Wellesley College and MIT
- AI for Science with Prof. Connor Coley, mentored by Samuel Goldman.
- Summer at the University of Oxford
- Clinical AI research mentored by Prof. David Clifton.
These experiences have shaped my interdisciplinary, people‑first approach to research and innovation.
Beyond the Lab
When I’m not debugging my models, you’ll find me playing tennis, doing pilates, spinning (indoor cycling), or collecting passport stamps (I’ve travelled to the Philippines, Singapore, Malaysia, Thailand, Indonesia, Vietnam, Japan, Korea, Australia, and France). Each adventure keeps me curious—and reminds me why building AI applications that works for everyone matters.
Let’s Connect 🤝
I love swapping ideas, exploring new collaborations, and mentoring budding researchers.
- Email: jiayixin [at] engineering [dot] upenn [dot] edu