Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
The paper introduces Dreaming in Code (DiCode), a framework that leverages foundation models to generate executable environment code variations for curriculum learning in open-ended environments. DiCode addresses the challenge of discovering learnable sequences in complex environments by "dreaming" or materializing code-level variations of the world to scaffold learning. Applied to the Craftax benchmark, DiCode achieves a 16% improvement in mean return over baselines and demonstrates success in late-game combat tasks, indicating the effectiveness of code-level environment design for curriculum control.
Foundation models can now dream up entire curricula by generating executable environment code, enabling agents to master complex, long-horizon tasks previously out of reach.
Open-ended learning frames intelligence as emerging from continual interaction with an ever-expanding space of environments. While recent advances have utilized foundation models to programmatically generate diverse environments, these approaches often focus on discovering isolated behaviors rather than orchestrating sustained progression. In complex open-ended worlds, the large combinatorial space of possible challenges makes it difficult for agents to discover sequences of experiences that remain consistently learnable. To address this, we propose Dreaming in Code (DiCode), a framework in which foundation models synthesize executable environment code to scaffold learning toward increasing competence. In DiCode,"dreaming"takes the form of materializing code-level variations of the world. We instantiate DiCode in Craftax, a challenging open-ended benchmark characterized by rich mechanics and long-horizon progression. Empirically, DiCode enables agents to acquire long-horizon skills, achieving a $16\%$ improvement in mean return over the strongest baseline and non-zero success on late-game combat tasks where prior methods fail. Our results suggest that code-level environment design provides a practical mechanism for curriculum control, enabling the construction of intermediate environments that bridge competence gaps in open-ended worlds. Project page and source code are available at https://konstantinosmitsides.github.io/dreaming-in-code and https://github.com/konstantinosmitsides/dreaming-in-code.