Open WorldDiscovery GitHub
← Research library
2026arXiv

Evolving with Open Worlds: Towards Recursive Self-Improvement for Future Forecasting

What this work does

DeepEvolve couples a live forecasting-task generator, an evidence-gathering forecaster, and an evolution engine. Forecasting trajectories support parameter training, reusable skill extraction, and generation of harder future tasks.

THE RESEARCH CONVERSATION

Discussion & reactions

Share a question, an insight, or a replication experience. Sign in with GitHub to join the conversation.

Open GitHub discussions ↗