Hi, I'm Angie! I am a Research Scientist at Apple. My research focuses on improving human-AI alignment by designing interfaces that communicate model representations and methods to shift them toward people's expectations.
Previously, I did my PhD at MIT CSAIL, where I was advised by Arvind Satyanarayan in the Visualization Group. My research was supported by the MIT John Jarve Fellowship and Apple Scholars in AI/ML PhD Fellowship, and I closely collaborated with Hendrik Strobelt at IBM Research. Before that, I received my SB and MEng in computer science from MIT, where I worked with Jim Glass and David Harwath in the Spoken Language Systems Group.
Recent News
I defended! Thank you to my committee: Arvind Satyanarayan, Martin Wattenberg, and Mitchell Gordon. 🎥 Watch my defense!
I am organizing the VISxAI workshop at IEEE VIS 2025. 📈 Stop by for a fireside chat with Adam Perer, Hendrik Strobelt, and Mennatallah El-Assady and a keynote by Martin Wattenberg.
This summer I am interning at Apple 🍎 working with Fred Hohman, Yannick Assogba, Donghao Ren, and Dominik Moritz.
I will be at CHI in Yokohama 🌸 presenting Abstraction Alignment.
I gave a guest lecture 📚 on AI interpretability in Brown's Sociotechnical Systems and HCI course.
I had a great time as a panelist at the NEMI workshop discussing how mechanistic interpretability can help keep AI safe and beneficial. 🤖
I am honored to receive an Apple Scholars in AI/ML PhD fellowship! 🏆
Publications
Semantic Regexes: Auto-Interpreting LLM Features with a Structured Language
International Conference on Learning Representations (ICLR) 2026
Abstraction Alignment: Comparing Model-Learned and Human-Encoded Conceptual Relationships
ACM Human Factors in Computing Systems (CHI) 2025
Compress and Compare: Interactively Evaluating Efficiency and Behavior Across ML Model Compression Experiments
IEEE Transactions on Visualization and Computer Graphics (VIS) 2024
VisText: A Benchmark for Semantically Rich Chart Captioning
Annual Meeting of the Association for Computational Linguistics (ACL) 2023
Outstanding Paper
Saliency Cards: A Framework to Characterize and Compare Saliency Methods
ACM Conference on Fairness, Accountability, and Transparency (FAccT) 2023
Shared Interest: Measuring Human-AI Alignment to Identify Recurring Patterns in Model Behavior
ACM Human Factors in Computing Systems (CHI) 2022
Best Paper Honorable Mention
Embedding Comparator: Visualizing Differences in Global Structure and Local Neighborhoods via Small Multiples
ACM Intelligent User Interfaces (IUI) 2022
Best Paper Honorable Mention