ModelsSelected
FrontierCS 2 上线 Kaggle：48 个任务测试智能体复现 CS 论文贡献
Why we picked it
Kaggle 搞了个狠基准：把论文贡献代码删掉，让 AI 在 5 小时内无网络复现并超越原作者指标，48 道题等你来看排行。
FrontierCS 2 是 Kaggle 上新推出的智能体基准，覆盖 13 个计算机科学领域的已发表论文，转化为 48 个任务。每个任务会把论文的贡献代码从开源实现中删除，智能体需要在作者自己的评估指标上重新超过原论文成绩。测试条件为无论文、无网络、限时 5 小时。目前榜单结果显示智能体仍有较大提升空间。
Original · Kaggle
Can AI agents contribute to CS research and development? 🤖
FrontierCS 2 is now live on Kaggle! This benchmark turns published papers across 13 areas of CS into 48 tasks: each paper's contribution is deleted from released code, and agents have to outperform it on the authors' own metrics.
No paper. No internet. Five hours.
Plenty of headroom left. See the results and leaderboard below 🧵
