About me

I am a second year Ph.D. student in the Allen School of Computer Science and Engineering at the University of Washington, advised by Yulia Tsvetkov. I do research in Natural Language Processing and Cognitive Modeling, with a growing emphasis on applying AI to improve healthcare and wellness outcomes.

I'm particularly interested in using computational methods to model cognitive processes, including how humans reason, communicate uncertainty, and make decisions in complex domains like healthcare. My long-term goal is to build socially and cognitively aligned NLP systems that support safer, more personalized, and equitable care.

Research interests: AI for Health, Safety & Reliability, Proactive Learning, Social Reasoning, and more!

Before grad school, I received my B.S. and M.S.E. at Johns Hopkins with majors in Computer Science, Cognitive Science (linguistics focus), and Applied Mathematics (statistics focus). I worked as a research assistant at the Center for Language and Speech Processing advised by Philipp Koehn and Kenton Murray.

Please contact me at stelli [at] cs.washington.edu if you are interested in my work!

  • Click here to view my CV (updated July 25)

     
  • Current Projects

    • design icon

      Proactive Reasoning

      I'm thinking about how to identify and proactively seek information using LLMs to improve model safety & reliability with statistical guarantee. How to make LLMs ask good questions? How do we model "intuition" in expert domains like medicine?

    • design icon

      Socially-Intelligent Personalization

      Modeling how different social groups express health concerns and interpret medical advice. Aiming to personalize AI systems for more equitable, culturally-aware health communication.

    News

    1. 2025-02

      Check out our new paper "PrefPalette: Personalized Preference Modeling with Latent Attributes" to model human preferences with cognitively grounded attributes for more accurate and explainable preference learning.

    2. 2025-06

      Presenting Spurious Rewards at Cohere Labs [YouTube] [Slides].

    3. 2025-06

      Prompt engineering can elicit similar behaviors in models as RLVR does. We show that "Spurious Prompt" can boost Qwen2.5-Math MATH-500 performance by 20% as well‼️ Check out our new blogpost "Spurious Rewards and Spurious Prompts."

    4. 2025-05

      Doing RLVR on incorrect and even random rewards can boost Qwen2.5-Math MATH-500 performance by 20%🤯 We explore how and why this happens in our new paper "Spurious Rewards: Rethinking Training Signals in RLVR" (blogpost).

    5. 2025-04

      Standard privacy evaluations focus mainly on surface-level lexical identifiers, but we show that semantic leakage is a more serious threat🚨. Check out our new paper "A False Sense of Privacy: Evaluating Textual Data Sanitization Beyond Surface-level Privacy Leakage."

    6. 2025-02

      Check out our new paper "ALFA: Aligning LLMs to Ask Good Questions A Case Study in Clinical Reasoning" by decomposing a complex goal into attributes and synthesizing paired data for preference learning.

    7. 2024-12

      Presenting MediQ at Neurips2024 in Vancouver 🍁.

    8. 2024-11

      Presenting ValueScope and Multilingual Abstention at EMNLP2024 in Miami 🌴.

    9. 2024-09

      Joining Meta FAIR as a visiting researcher.

    Experience

  • Click here to view my CV (updated July 25)

     
  • Education

    1. University of Washington

      2023 — present | Seattle, WA

      Ph.D. in Computer Science and Engineering

      Advised by Yulia Tsvetkov.

    2. Johns Hopkins University

      2022 — 2023 | Baltimore, MD

      M.S.E. in Computer Science with Human Language Technology Concentration

      Advised by Philipp Koehn and Kenton Murray.

      Thesis: Learning from Gibberish: Code-Mixing Data Augmentation for Sentiment Analysis

    3. Johns Hopkins University

      2019 — 2022 | Baltimore, MD

      B.S. in Applied Mathemtics and Statistics

      Advised by Philipp Koehn and Ed Scheinerman.

      Other Majors: Computer Science, Cognitive Science (linguistics focus)

      Minor: Mathematics

    4. Stanford Online High School

      2018 — 2019 | Palo Alto, CA

      Dual enrollment program with a focus in advanced mathematics

    5. Robert Louis Stevenson School

      2016 — 2019 | Pebble Beach, CA

      Awards & Leadership: Cum Laude Society, USABO Semifinalist, USAMO Qualified, Bausch & Lomb National Science Award, Math Madness Silver Medalist, Math Team Captain, Spanish National Honor Society, Varsity Volleyball

    Teaching/TA Experience

    1. Ethics in AI: Teaching Assistant

      2025 Winter

      CSE 582

    2. Introduction to Statistics: Teaching Assistant

      2020 Spring, 2021 Fall, 2022 Spring, 2023 Spring

      EN.503.430 (undergrad) & EN.503.630 (grad) & EN.503.431 (honors)

    3. Artificial Intelligence: Course Assistant

      2023 Spring

      EN.601.464 (undergrad) & EN.601.664 (grad)

    4. Human-Computer Interaction: Course Assistant

      2022 Fall

      EN.601.490 (undergrad) & EN.601.690 (grad)

    5. Computer Ethics: Head Course Assistant

      2022 Summer

      EN.601.104

    6. Intermediate Programming: Course Assistant

      2020 Spring, 2021 Fall, 2022 Spring

      EN.601.220

    Work Experience

    1. Meta FAIR: Visiting Researcher

      2024 - Current | Seattle, WA

      Advised by Asli Celikyilmaz.

      Working on Social Alignment on the SAGE Team at Meta FAIR.

    2. Yext: Software Engineering Intern

      2022 Summer | Arlington, VA

      Integrated client data to Yext platform for real-time site information updates using Go.

      Created a Figma Style Picker to improve developer workflow and scalability using ReactJS.

    3. Michigan State University: Research Intern

      2021 Summer | East Lansing, MI

      Advised by Wolfgang Banzhaf.

      Designed and implemented novel GP algorithm for LLVM compiler flag optimization (20%).

      Published work at GECCO; second author of GP paper; first author of GI paper.

    4. Bytedance AI Lab: Research Intern

      2020 Summer | Beijing, China

      Trained neural networks for text normalization in text-to-speech tasks.

      Implemented statistical information-retrieval algorithms for theme clustering and complexity ranking for TikTok videos.

    5. Johns Hopkins Language and Cognition Lab: Research Assistant

      2020 - 2022 | Baltimore, MD

      Advised by Barbara Landau.

      Investigated developmental spatial cognition using Lego Block building.

      Created ML model for movement prediction and stability analysis using motion sensor data.

    Publications

    Below is a list of projects for which I was very involved in (lead/co-lead/contributed significantly). For a more comprehensive list of papers, check out my Google Scholars page. I also try to record the time that I spent on each project in case anyone finds it helpful!

    Photography