Born a Transformer -- Always a Transformer? On the Effect of Pretraining on Architectural Abilities. Born a Transformer -- Always a Transformer? On th

Born a Transformer -- Always a Transformer? On the Effect of Pretraining on Architectural Abilities w/ Yash Sarrof & Aleksandra Bakalova of Saarland University

Speakers: Aleksandra Bakalova, Yash Sarrof

Arxiv_id: 2505.21785v3