Ai system improves itself by learning in a changing environment
Evolving with Open Worlds: Towards Recursive Self-Improvement for Future Forecasting
Computation and Language
Summary
Most current AI systems don’t learn from their own experience once they are made; they only get better through external updates. The authors introduce DeepEvolve, an AI agent that keeps learning and improving by interacting with a changing world and using what it learns to get better at predicting future events. This system participated in a public prediction challenge and ranked first for several weeks, showing it can improve itself effectively. Their work suggests future AI should be designed to grow from ongoing experience in real-world conditions, rather than being fixed after training.
What this means in practice
- •For forecasting teams: Develop AI forecasters that improve predictions over time by adapting to changing real-world conditions.
- •For game developers: Create adaptive game agents that continuously learn from player actions within dynamic game worlds.
Authors
Zhipeng Chen, Yunsen Liang, Yanxi Chen, Hongkang Zhao, Lingyun Feng, Ruiyang Ren, Guoxin Chen, Jia Deng, Feng Chang, Chuan Hao, Ran Tao, Jian Yang, Ji-Rong Wen, Bryan Dai, Wayne Xin Zhao
Abstract
Most contemporary artificial intelligence (AI) systems are released with frozen parameters and improve through external updates, whereas an ideal system would keep improving from its own experience, a capacity we call recursive self-improvement (RSI). Yet studying RSI requires an environment that is continuously changing, objectively verifiable, and naturally grounded, posing a fundamental challenge for agent evolution. Here we introduce DeepEvolve, a sustainable evolution agent system grounded in future event prediction, in which agents continuously interact with a dynamically evolving environment and improve by internalizing and externalizing the experience gained through these interactions. A living environment synthesizes verifiable tasks, and a recursive-evolution engine turns the reasoning trajectory behind each prediction back into capability: decoupling search from prediction adapts parameter-level evolution to the contamination-prone live environment, while distilling reusable skills lets the agent evolve from the limited live trajectories such an environment affords. On the external FutureX leaderboard, which evaluates predictions against real-world outcomes and is updated weekly, our self-improved forecaster ranked first across all systems over three consecutive weeks, achieving up to a 195% improvement in rating. On our synthesized evaluation, it ranked among the top three across four base LLMs, outperforming the strongest baseline by up to 10.7 Bradley--Terry points. More broadly, these results suggest that a system built for self-improvement is best understood as an agent that co-evolves with a genuinely living, open-ended environment, pointing to a shift in AI from training static models on fixed data toward cultivating agents that grow stronger from their own experience, with such an open world as the indispensable basis of that growth.