PAP
The Scaling Properties of Implicit Deductive Reasoning in Transformers
May 5, 2026
Abstract
We investigate the scaling properties of implicit deductive reasoning over Horn clauses in depth-bounded Transformers. By systematically decorrelating provability from spurious features and enforcing algorithmic alignment, we find that in sufficiently deep models with a bidirectional prefix mask, implicit reasoning approaches explicit CoT performance across graph topologies and problem widths, though CoT remains necessary for depth extrapolation.
Select text to highlight · click a highlight to remove · saved in this browser only
Authors
Enrico Vompa, Tanel Tammet
Your notes (browser-local)
savedCross-links
arXiv:2605.04330