模型多源确认

开发者实测 GPT-6 Astra 灵巧操作:零样本转笔、复原魔方、用锤子

源:https://t.co/2MG5no1UMg

精选理由

有人用 GPT-6 Astra 让机器手零样本转笔、玩魔方,模拟训练后还会用锤子,代码放在 GitHub 上。

开发者 Jianglong Ye 在 X 上分享了 GPT-6 Astra 的灵巧操作实测结果。该模型无需针对具体任务微调,即可零样本完成转笔、复原魔方、使用锤子等精细手部操作。进一步的提升来自模拟训练,模型在仿真环境中持续学习并改进这些技能。他同时公开了 GitHub 仓库,并指出真正的目标是让这种灵巧能力在真实世界中进化。

原文 · AI Will

源:https://t.co/2MG5no1UMg

源: x.com/jianglong_ye/s… Jianglong Ye @jianglong_ye Evolving dexterity with GPT-6 Astra 🖐️ Been trying Astra recently. Its zero-shot dexterous manipulation is already quite surprising. More interesting is seeing it learn and improve through simulation training, from pen spinning and Rubik's Cube to hammer use. The real goal would be to evolve this dexterity in the real world. github.com/jianglongye/de… c Your browser does not support the video tag. 🔗 View on Twitter 🔗 View Quoted Tweet 💬 0 🔄 0 ❤️ 0 👀 249 ⚡