Anna Korhonen

Latest

• Self-Improving World Modelling with Latent Actions
• Bolmo: Byteifying the Next Generation of Language Models
• Bootstrapping Action-Grounded Visual Dynamics in Unified Vision-Language Models