Demonstrates that symbolic event primitives (like Schank's Conceptual Dependency) can be 'rediscovered' by neural networks purely through compression pressure.
March 30, 2026
Original Paper
Do Neurons Dream of Primitive Operators? Wake-Sleep Compression Rediscovers Schank's Event Semantics
arXiv · 2603.25975
The Takeaway
This bridges the gap between connectionism and symbolic AI by showing that Minimum Description Length (MDL) optimization naturally recovers human-interpretable event structures. It suggests that complex commonsense reasoning can be learned unsupervised by seeking the most compressed representation of state changes.
From the abstract
We show that they do. Schank's conceptual dependency theory proposed that all events decompose into primitive operations -- ATRANS, PTRANS, MTRANS, and others -- hand-coded from linguistic intuition. Can the same primitives be discovered automatically through compression pressure alone?We adapt DreamCoder's wake-sleep library learning to event state transformations. Given events as before/after world state pairs, our system finds operator compositions explaining each event (wake), then extracts