精选80Rohan Paul论文研究
斯坦福与东北大学推出“Git for AI agents”
Stanford and Northeastern built "Git for AI agents."
Stanford and Northeastern built "Git for AI agents."
斯坦福和东北大学构建了“面向AI代理的Git”。
Solves a common problem with existing agent frameworks where record only transcripts and environment snapshots, so rewinding recovers files but not the live process or provider cache.
解决了现有代理框架中的一个常见问题，即只记录转录和环境快照，因此回滚可以恢复文件，但无法恢复活动进程或提供者缓存。
- It commits an agent's running process and filesystem together
- makes an agent's execution revertible like Git. Check out any past commit and resume from that exact state.
- turns every model action, tool call, and environment change into a typed commit.
- 2 coding agents splitting one repo normally score worse than a single agent doing both jobs alone: 28.8% versus 57.2%. Shepherd closes 91% of that gap by letting a supervisor watch both runs live and rewind one before a bad write lands, which is the clearest proof that reversible traces buy real performance, not just cleaner debugging.
- 它将代理的运行进程和文件系统一起提交。
- 使代理的执行像Git一样可回滚。检出任何过去的提交，并从该确切状态恢复。
- 将每个模型动作、工具调用和环境变化都转化为类型化提交。
- 两个编码代理分割一个仓库通常比单个代理单独完成两项工作的得分更低：28.8%对比57.2%。Shepherd通过让监督者实时观察两次运行，并在错误写入发生前回滚其中一个，缩小了91%的差距，这最清楚地证明了可逆追踪带来了真正的性能提升，而不仅仅是更清晰的调试。
更进一步：量化金融体系
看懂新闻只是起点——沿量化金融路径，把它变成能交付的工程能力