ERM Fixes Causal Rung Collapse in LLMs

有什麼變化

Formalizes rung collapse as lack of gradient for P(Y|do(X)) vs P(Y|X)

為什麼重要

Addresses core reasoning flaws in LLMs, enabling better generalization and steerability. Could prevent entrenchment in production models, improving reliability across domains. Inverse scaling in steerability highlights need for targeted causal fixes.

下一步行動

Prioritize whether this update affects your current workflow this week.