精选理由
NVIDIA 让智能体自己搭环境教模型，Qwen3-VL-2B 准确率从 25% 飙到 96.9%，还能自动想下一步实验，挺有意思。
NVIDIA 展示一个编码智能体，利用 NeMo RL 和 NeMo Gym 框架，在有限时间内自主构建训练环境并指导 Qwen3-VL-2B 视觉模型学习数彩色星星。模型准确率从 25% 跃升至 96.9%。智能体还能主动提出下一项实验方案，研究员全程仅进行方向性指导。
原文 · NVIDIA AI
We gave a coding agent a goal and a time budget: build a training environment and teach a vision mod...
We gave a coding agent a goal and a time budget: build a training environment and teach a vision model to count colored stars. Using autoresearch with NeMo RL, NeMo Gym, and reusable skills, the agent set up, trained and evaluated the model while the researcher steered the work. Qwen3-VL-2B went from 25% to 96.9% accuracy, and the agent even proposed the next experiment on its own. 💬 11 🔄 5 ❤️ 51 👀 2695 📊 19 ⚡