aifollow.news
返回 Rohan Paul
Rohan Paul· @rohanpaul_ai · X· · 原发布时间 AI 评分54

LongChang 用改写物理定律的模拟世界测试 AI 自主发现能力

自动核验发布 · 本文由系统生成并完成证据核验,未经人工审稿。

AI 导读

LongChang 团队让 12 个大语言模型探索 14 个物理定律被改写的模拟世界。团队报告,GPT-6-Astra 的发现率为 16.7%,其余 11 个模型为 0%;将探索预算增加至 5 倍也未带来改善。这些是团队公布的评测结果。

原帖引用 · 原文

@rohanpaul_ai

@XuXander24218 Grad students in physics education could use these scenes to teach controlled experiments

回复Xander Xu@XuXander24218
Can AI discover physical laws that humanity has never known? Excited to share LongChang (龙场), our team's work evaluating frontier AI models on autonomous scientific discovery. Named after the place where Chinese philosopher Wang Yangming (王阳明) attained enlightenment, LongChang challenges 12 LLMs to explore 14 simulated worlds with fundamentally altered physical laws. 📊 Results: - GPT-6-Astra: 16.7% discovery rate - All other 11 models: 0% - 5× more exploration budget brought no improvement. Can AI discover what it doesn't even know to ask? 🌐 Try our interactive demo directly on our website! We'd love to hear what you think. 📄 Paper: https://t.co/kLzkxN5g8r 💻 Code: https://t.co/tzfNqEdoxC first author @ZhenYang_CS
在 X 查看这条帖子 ↗

来源:Rohan Paul · x.com

论文
发现内容有误?提交纠错