collaborations.txt

I am broadly interested in collaborating on:
- alignment
- neurosymbolic ai
- causality

Some specific research directions are:
- neurosymbolic causal discovery to learn lifted rule representations
- causal abstraction in world models for long horizon planning
- intervention semantics for latent variable representations in world models
- neurosymbolic attention mechanisms
- the limits of preference distribution learnability
- hci dynamics for alignment

Reach out below if you're keen to chat.