#artificial intelligence
May 2026
One Model, Two Roles: Emergent Specialization in a Shared Recurrent Transformer
The results show that, in a two-state recurrent setting, a clear state-identity signal can induce stable, related functional roles inside a shared-parameter recurrent Transformer.
Jucheng Shen, Barbara Su, Anastasios Kyrillidis
· arXiv.org · 1 citation