发表机构
Columbia University; Google（哥伦比亚大学; 谷歌）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对终端智能体探索中环境状态分支恢复的难题，提出StateFork控制平面与Waypoint检查点基底，实现观测等价且物理高效的分支，在Terminal-Bench上提升任务完成率并加速探索。
AI 中文摘要
AI智能体通过探索多条轨迹来提高任务成功率，但对于计算机使用型智能体，每条轨迹都会修改外部环境状态。仅当恢复在观测上等价（即未来动作产生相同观测）时，从中间点分支才是正确的；且仅当创建、恢复和丢弃分支状态在物理上高效时，分支才具有实用性。我们针对使用终端的智能体研究该问题，此类任务会修改文件、shell上下文、运行中的进程和本地服务。我们提出StateFork，一个逻辑控制平面，将探索策略与物理状态物化解耦，在多种执行基底上暴露会话、命令、快照、恢复和清理操作。我们还构建了Waypoint，一个用于终端执行会话的检查点/恢复基底，结合了文件系统分层、进程检查点以及持久的终端兼容命令会话。StateFork和Waypoint共同通过结合样本高效的搜索与对正确执行状态的高效恢复，提升了终端智能体的探索能力。在Terminal-Bench上，通过StateFork和Waypoint进行基于分支的探索，在相同访问节点预算下相比pass@20提升了任务完成率，且使用Waypoint相比其他执行基底实现了高出26%的任务准确率，同时探索速度最高提升70%。这些结果表明，观测等价且物理高效的执行会话是探索性AI智能体的关键系统抽象。
英文摘要
AI agents improve task success by exploring multiple trajectories, but for computer-use agents each trajectory modifies external environment state. Branching from an intermediate point is correct only when restoration is observation-equivalent - future actions produce the same observations - and practical only when creating, restoring, and discarding branch states is physically efficient. We study this problem for terminal-using agents, where tasks modify files, shell context, running processes, and local services. We introduce StateFork, a logical control plane that separates exploration policies from physical state materialization, exposing sessions, commands, snapshots, restores, and cleanup over multiple execution substrates. We also build Waypoint, a checkpoint/restore substrate for terminal execution sessions that combines filesystem layering, process checkpointing, and a persistent terminal-compatible command session. Together, StateFork and Waypoint improve terminal-agent exploration by combining sample-efficient search with efficient restoration of the right execution state. On Terminal-Bench, branch-based exploration through StateFork and Waypoint improves task completion over pass@20 at the same visited-node budget, and using Waypoint achieves 26% higher task accuracy than other execution substrates while completing exploration up to 70% faster. These results show that observation-equivalent, physically efficient execution sessions are a key systems abstraction for exploratory AI agents.
