Совместная разработка Cognition и Стэнфорда, нацеленная на революцию в написании кода. Kevin-32B не просто выдает текст, а обучается в процессе итераций, имитируя работу реального программиста. Эта ИИ-модель использует обучение с подкреплением для достижения максимальной точности в генерации кода.
Coding is an iterative process – you write a program, execute it, evaluate the results, and refine your code based on the feedback. Recent advances in LLM for code generation have tried to incorporate this process at inference-time, using methods like parallel sampling. While these methods are effective, they rely on search without actual learning — the model weights are frozen. We explore reinforcement learning in a multi-turn setting, using intermediate feedback from the environment, and masking model thoughts to avoid exploding context over multiple turns. Our model, Kevin-32B = K(ernel D)evin, outperforms frontier reasoning models on kernel generation. Moreover, our results show that multi-turn training makes the model more effective at self-refinement compared to single-turn training.