Abstraction Alignment works on datasets too!

Medical experts analyzed clinical dataset abstractions, uncovering issues like overuse of unspecified diagnoses.

This mirrors real-world updates to medical abstractions — showing how models can help us rethink human knowledge.

Language models often prefer specific answers even at the cost of performance.

But Abstraction Alignment reveals that the concepts an LM considers are often abstraction-aligned, even when it’s wrong.

This helps separate surface-level errors from deeper conceptual misalignment.

And we packaged Abstraction Alignment and its metrics into an interactive interface so YOU can explore it!

🔗https://vis.mit.edu/abstraction-alignment/

Abstraction Alignment compares model behavior to human abstractions.

By propagating the model's uncertainty through an abstraction graph, we can see how well it aligns with human knowledge.

E.g., confusing oaks🌳 with palms🌴 is more aligned than confusing oaks🌳 with sharks🦈.

Interpretability identifies models' learned concepts (wheels 🛞).

But human reasoning is built on abstractions — relationships between concepts that help us generalize (wheels 🛞→ car 🚗).

To measure alignment, we must test if models learn human-like concepts AND abstractions.

#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