We introduce Evolutionary Ensemble (EvE), a decentralized framework that organizes existing, highly capable coding agents into a co-evolving system for algorithmic discovery. Rather than reinventing the wheel, EvE fixes the base agent substrate and focuses entirely on evolving the guidance and skills that dictate agent behaviors. When applied to a research bottleneck in In-Context Operator Networks (ICON), EvE autonomously discovered a robust mechanism that enables reliable context length generalization.
A single GPT or Claude subscription is sufficient, with no additional API fees. Third-party APIs are also supported.

Zongmin Yu, Liu Yang^†

We introduce In-Context Modeling (ICM), a retrain-free paradigm that infers physical relationships directly from observational fields. Rather than encoding system-specific behavior in fixed parameters, ICM assimilates measurements as physical context and performs inference through a single forward pass. Trained in a physics-informed, label-free manner using governing equations, a single model generalizes across unseen materials, geometries, and loading conditions.

Lingfeng Li^*, Zhuoyuan Li^*, Shun Li^*, Kaixin Zhan, Huajian Gao^†, Changqing Chen^†, Liu Yang^†

Can we build a single large model for a wide range of PDE-related scientific learning tasks? We propose in-context operator learning framework with corresponding model In-Context Operator Network (ICON). A singe ICON model can act as an operator learner for a diversified type of differential equation problems, including forward and inverse problems of ODEs, PDEs, and mean-field control problems, without any fine-tuning.

Liu Yang, Siting Liu, Tingwei Meng, Stanley J. Osher^†