Paper: Transformers Converge to Invariant Algorithmic Cores Across Models
A new paper by Joshua S. Schiffman introduces Algorithmic Core Extraction (ACE), showing that independently trained Transformers converge to compact, invariant subspaces that are necessary and sufficient for a task and can be extracted and manipulated across models.
2026-11-20 ~ 2026-11-20 · 2 related posts
- Paper: Transformers converge to invariant algorithmic cores across training runs — pickover · 2026-11-20
- ACE extracts invariant algorithmic cores shared across independently trained transformers — burny_tech · 2026-11-20