In the news
Surprisal Theory is Tautological (without Rational Grounding)
arXiv cs.AI · Published · 3 min read
In 30 seconds
- What happened
- Surprisal theory, which links human reading difficulty to language model uncertainty, is logically tautological without additional constraints on the model itself.
- Why it matters
- Matters for NLP researchers building cognitive models and psycholinguists validating theories against human processing data.
- Watch out
- The paper argues surprisal theory needs external constraints from cognitive theory, not just corpus-fit improvements, to make falsifiable predictions about human behavior.
- language model
The patterns behind this
Each one covers how the technique works, when it earns its cost, and where it breaks.
The Agent Architect
One pattern, one tradeoff, one production failure story. A short weekly briefing for people building agentic systems.
Weekly email, one-click unsubscribe. We only use your address to send the briefing.