Angie Boggust

@angieboggust.bsky.social

MIT PhD candidate in the VIS group working on interpretability and human-AI alignment

I’ll be at #CHI2025 🌸 If you are excited about interpretability and human-AI alignment — let’s chat! And come see Abstraction Alignment ⬇️ in the Explainable AI paper session on Monday at 4:20 JST

Angie Boggust@angieboggust.bsky.social · 2y ago

#CHI2025 paper on human–AI alignment!🧵 Models can learn the right concepts but still be wrong in how they relate them. ✨Abstraction Alignment✨evaluates whether models learn human-aligned conceptual relationships. It reveals misalignments in LLMs💬 and medical datasets🏥. 🔗 arxiv.org/abs/2407.12543

An overview of Abstraction Alignment, including its authors and links to the paper, demo, and code.