activity
20242026
most citedLearning interpretable positional encodings in transformers depends on initialization

1 citations · 1 across the 6 of their papers we have counts for

collaborators
Showing 2026Show all

2 papers · 1 filter

cs.LG2026

Unrolled Flow Models for Reasoning

Faissal Izermine, Hanru Bai, Oscar Davis +1

Flow matching enables language generation in few steps, but whether additional integration steps improve reasoning remains unclear. We prove that a flow parameterized by a two-laye…

cs.LG2026

Universal Transformers for Circuit Computations: Perfect Length Generalization in Tiny Transformers

Takuya Ito, Ruchir Puri, Murray Campbell +1

Learning generalizable algorithmic computations remains a challenge for neural networks, as reflected in persistent failures on compositional and length generalization benchmarks.…