Author

Enrico Vompa

2 papers indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

#artificial intelligence May 2026

The Scaling Properties of Implicit Deductive Reasoning in Transformers

It is found 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.

Enrico Vompa, T. Tammet · arXiv.org · 1 citation

#artificial intelligence Preprint Sep 2026

High-probability guarantees for linear accessibility in feature superposition

This work characterize the asymmetry between active and inactive interference and the trade-off between interference and observation-noise budgets and introduces IHT-SAE, which uses learned iterative refinement to improve feature recovery beyond the limits of linear availability.

Enrico Vompa · 0 citations