回归开发状态：完整沙盒快照何时对编码智能体有帮助
Returning to Development States: When Full Sandbox Snapshots Help Coding Agents
AI总结：
该研究针对编码智能体失败后从头重试丢弃中间状态的问题，提出基于 Daytona 完整快照的 Go-Explore 风格搜索，发现其在长周期编码工作跨多个服务时能提升解决率。
摘要：
编码智能体在失败前往往能取得有用进展，但从头重试会丢弃过程中产生的机器状态，这使得失败恢复成为一个有趣的搜索问题：若中间环境状态保留了有意义的进展，从这些状态恢复并分支可能比从初始状态反复重启表现更好。我们在完整的 Daytona 快照上实现了 Go-Explore 风格的搜索，发现更高且更高效的解决率主要取决于存储库之外的进展，如数据库、服务及其他运行时状态。我们的结果表明，当长周期编码工作跨越多个无法通过单一差异重建状态的服务时，完整快照搜索最为有效。
英文摘要：
Coding agents often make useful progress before they fail, but retry-from-scratch discards the machine state created along the way. This makes failure recovery an interesting search problem: if intermediate environment states preserve meaningful progress, restoring and branching from them could outperform repeatedly restarting from the initial state. We implement Go-Explore-style search over full Daytona snapshots and find that higher and more efficient solve rates depend mainly on progress that sits outside the repository, such as databases, services, and other runtime state. Our results suggest that full-snapshot search is most effective when long-horizon coding work spans multiple services whose state cannot be reconstructed from a single diff.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
