Helen Jin

Helen Jin

Research Interests: Explainable AI (XAI) × Education, Trustworthy AI, Human & Machine Understanding, Cognition & Learning, Education β˜€οΈ

Welcome! I am a Ph.D. candidate in Computer and Information Science (CIS) at the University of Pennsylvania, where I am extremely grateful to be co-advised by Prof. Boon Thau Loo and Prof. Seiji Isotani. I am also affiliated with Penn NLP, ASSET, BrachioLab, and Penn GSE Living Laboratory for Human Agency and AI. Previously, I earned my B.A. from Columbia University β™”, with a double major in Mathematics and Computer Science and a concentration in East Asian Languages and Cultures (EALAC).

My research lies at the intersection of trustworthy AI, human and machine understanding, and education, guided by two fundamental questions:

  1. How can we establish and sustain principled trust between humans and AI systems?
  2. How can we assess what humans and machines truly understand?

I develop methods to evaluate the reliability, explainability, and reasoning of AI systems, examining whether they are correct for the right reasons and aligned with domain expertise, including in scientific reasoning and claim verification. More recently, my research has naturally expanded into education, more specifically AI in Education (AIED), motivated in part by my lifelong passion for teaching, my many (and hopefully more!) experiences as an educator, and growing questions about how teaching and learning must evolve in the age of generative AI. In particular, I am interested in applying machine learning and explainable AI (XAI) techniques to education problems. My current work focuses on educational assessment and building explainable models of student success, aiming to better understand student learning and outcomes, as well as help educators provide more effective support as early as possible.

πŸ’‘ I firmly believe AI should ultimately serve humanity, not take away its joy. As we figure out how AI fits into our society (as these are still the early days!), it is pivotal that we develop both a deep understanding of the humanities and a rigorous technical grasp of the technology itself.

Please feel free to reach out to discuss potential collaborations and opportunities.

News & Highlights

Show moreShow less

Selected Publications

* equal contribution. Full list on Google Scholar.

  • Probabilistic Stability Guarantees for Feature Attributions Helen Jin*, Anton Xue*, Weiqiu You, Surbhi Goel, Eric Wong NeurIPS 2025 [arXiv] [Paper] [Poster] [Blog]
  • The FIX Benchmark: Extracting Features Interpretable to eXperts Helen Jin*, Shreya Havaldar*, Chaehyeon Kim*, Anton Xue*, Weiqiu You*, et al., Eric Wong DMLR 2025 [arXiv] [Paper] [Blog] [Website] You get to see my dog, Asta! 🐢
  • Probabilistic Soundness Guarantees in LLM Reasoning Chains Weiqiu You, Anton Xue, Shreya Havaldar, Delip Rao, Helen Jin, Chris Callison-Burch, Eric Wong EMNLP 2025 [arXiv] [Paper]

Teaching Experience

“Thus it is proper to the role of the scientist that he not merely find the truth and communicate to his fellows, but that he teach, that he try to bring the most honest and most intelligible account of new knowledge to all who will try to learn.”

— J. Robert Oppenheimer

University of Pennsylvania

  • Summer 2026 Guest Lecturer, CIS 1210: Data Structures and Algorithms
  • Spring 2026 Teaching Assistant, CIS 5200: Machine Learning Led the design and rollout of LLM-assisted oral assessments for ~200 students, Penn's first large-scale replacement of written homework grading with oral assessments.
  • Fall 2025 Teaching Assistant, EAS 2030: Engineering Ethics
  • Fall 2024 Teaching Assistant, CIS 3333: Mathematics of Machine Learning
  • Summer 2023 Teaching Assistant, Girls Who Invest Summer Intensive Program @ The Wharton School
  • Fall 2021 Teaching Assistant, CIS 4210/5210: Artificial Intelligence Penn's largest course (~500 students); managed two sections simultaneously.

Lehman College, City University of New York

  • Summer 2020 Data Science Mentor-Instructor, Pathways to Student STEM Success Program (PTS3)

Columbia University

Teaching Assistant for 10 course offerings across 6 courses as an undergraduate.

Show moreShow less
  • Summer 2020 Teaching Assistant, Natural Language Processing
  • Spring 2020 Teaching Assistant, Natural Language Processing
  • Fall 2019 Teaching Assistant, Natural Language Processing
  • Summer 2019 Teaching Assistant, Calculus III
  • Summer 2019 Teaching Assistant, Calculus IV
  • Spring 2019 Teaching Assistant, Linear Algebra
  • Fall 2018 Teaching Assistant, Calculus III
  • Spring 2018 Teaching Assistant, Calculus IV
  • Fall 2017 Teaching Assistant, Calculus I
  • Fall 2017 Teaching Assistant, Calculus-Based Intro to Statistics

Completed the CETLI Teaching Certificate at Penn.

Academic Service

Research Community

  • Reviewer ACL, NAACL, NeurIPS, DMLR, SIGCSE, China National Conference on Computational Linguistics (CCL)
  • Spring 2023 Organizer, Computational Linguistics Seminar (CLunch), University of Pennsylvania
  • 2023, 2025 Conference Volunteer, NAACL 2023 and NeurIPS 2025

Department Service

University Service

  • 2021–2025 Executive Board, Graduate and Professional Student Assembly (GAPSA) Director of Cultural Programming (2024–25), Vice President of Operations (2022–23), Director of Logistics (2021–22)
  • 2021–2023 Graduate Student Representative, University Council; University Committee on Honorary Degrees; University Committee on Academic Affairs; Student Advisory Group for the Environment (SAGE)
  • 2021–2023 Penn Representative, Ivy+ Summit Leadership Conference

Outreach

Honors & Awards

Asta & Co.

My dog, Asta 🐢 (who also makes a cameo in the FIX benchmark!), and my cat, Noelle πŸˆβ€β¬›.

I work to provide for their royal highnesses. πŸ‘‘

Fun fact: Every pedigreed French cat gets an official name whose first letter encodes its birth year, cycling through the alphabet each year and skipping K, Q, W, X, Y, and Z.[1] A Chartreux born in 2023 should have a name starting with U, so Noelle, born in 2023, is a little rebel. 😼

Photos coming soon! 🐾

More pictures to come! 🐾