让AI agent长期驻守科研环境，实测其持续工作能力与产出效率。
Persistent AI Agents in Academic Research: A Single-Investigator Implementation Case Study
- 构建持久科研代理环境，集成记忆、工具与安全规则
- 累计运行579.7小时，生成627个完成任务，82.9%为缓存读取
- 提出以成果为单位的评估框架，适合长期研究自动化者
背景：大语言模型通常作为模型、基准或短对话进行评估。但当智能体被嵌入真实科研环境中并具备持久记忆、本地文件、外部工具、定时任务、角色委派和明确安全协议时，其表现尚不明确。方法：2026年1月31日至5月25日开展结构化自观察案例研究，分析对象为持久人机协作环境：研究人员、智能体运行时、记忆层、工具、仓库、定时任务、专用角色与治理规则。采用PARE-M（持久智能体研究环境测量）框架评估架构、使用率、成果产出、资源消耗、可复现性及治理情况。结果：可恢复的主代理遥测数据包含75,671条去重记录，覆盖96个活跃日，用户角色消息8,059条，助手角色消息23,710条。工作区含502个记忆相关文件、17个配置的代理目录和57个技能文件。系统活跃时间579.7小时（30分钟断点估算）。基于记忆的记录识别出482次输出代理事件和889次失败、验证、修正或协议代理事件。严格的5月2026年轨迹子集捕捉到627个模型完成事件，共记录7395万词元，其中82.9%为缓存读取。结论：该流程以缓存为主导，表明持久智能体环境可能使经济单位从每词元成本转向每成果成本。未来评估应采用成果级计量单位、可复现解析规则、纠错分类体系及独立治理事件编码。
原文摘要 · Abstract (English)
Background: Large language models are typically evaluated as models, benchmarks, or short conversational episodes. Less is known about what happens when an agent is embedded persistently in a real academic research environment with durable memory, local files, external tools, scheduled routines, delegated roles, and explicit safety protocols. Methods: A structured self-observed implementation case study was conducted from January 31 to May 25, 2026. The unit of analysis was the persistent human-agent environment: researcher, agent runtime, memory layer, tools, repositories, scheduled jobs, specialized agent roles, and governance rules. Outcomes were organized using PARE-M (Persistent Agentic Research Environment Measurement), a measurement framework covering architecture, utilization, artifact production, resource use, reproducibility, and governance. Results: Recoverable main-agent telemetry contained 75,671 de-duplicated records across 96 active days, with 8,059 user-role and 23,710 assistant-role messages. The workspace included 502 memory-related files, 17 configured agent directories, and 57 skill files. Active system time was 579.7 hours (30-minute capped-gap estimate). Memory-derived records identified 482 output-proxy events and 889 failure, verification, correction, or protocol-proxy events. A strict May 2026 trajectory subset captured 627 model-completed events and 73.95 million recorded tokens, of which 82.9% were cache reads. Conclusions: The workflow was cache-dominant, suggesting that persistent agentic environments may shift the economic unit from cost per token to cost per completed artifact. Future evaluations should use artifact-level denominators, reproducible parsing rules, correction taxonomies, and independent coding of governance events.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。