paper-with-me

홈 › Papers

DeltaBox: Scaling Stateful AI Agents with Millisecond-Level Sandbox Checkpoint/Rollback

2026-05-21 · Yunpeng Dong, Jingkai He, Shiqi Liu, Yuze Hou, Dong Du, Zhonghu Xu, Si Yu, Baochuan Yang, Yubin Xia, Haibo Chen arxiv

LLM-powered AI agents require high-frequency state exploration (e.g., test-time tree search and reinforcement learning), relying on rapid checkpoint and rollback (C/R) of the complete sandbox state, including files and process state (e.g., memory, contexts, etc.). Existing mechanisms duplicate the entire state, causing hundreds of milliseconds to seconds of latency per C/R, which severely bottlenecks deep search and large-scale fan-outs. This paper observes that subsequent checkpoints in AI agents are highly similar. Therefore, instead of full duplication, a sandbox should only duplicate the changes between consecutive checkpoints (Key Insight). However, it is non-trivial to realize the idea, mainly due to the missing OS supports. This paper proposes a new OS-level abstraction, DeltaState, to enable the change-based transactional C/R for AI agents with two co-designed OS mechanisms. First, DeltaFS enables change-based filesystem C/R by organizing the file states into layers and dynamically freezing the writable layer and inserting a new one during checkpoint, reducing file updates to copy-on-write, and making rollback a simple layer switch. Second, DeltaCR enables change-based process state C/R using incremental dumps, and accelerates rollback by bypassing traditional pipelines to directly fork() from a frozen template process. We then present DeltaBox, a novel agent sandbox achieving millisecond level C/R through the two new mechanisms. Evaluations on SWE-bench and RL micro-benchmarks show DeltaBox completes checkpoint and rollback in millisecond-level latency (14ms and 5ms, respectively), empowering agents to explore substantially more nodes under fixed time budgets.

📄 PDF Abstract BibTeX arXiv:2605.22781

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement Learning

Similar Papers 제목 키워드 기반

Scalable Environments Drive Generalizable Agents

2026-05-18 · Jiayi Zhang, Fanqi Kong, Guibin Zhang, Maojia Song 외 arxiv

Generalizable agents should adapt to diverse tasks and unseen environments beyond their training distribution. This position paper argues that such generalization requires environment scaling: expanding the distribution …

MuSEAgent: A Multimodal Reasoning Agent with Stateful Experiences

2026-03-29 · Shijian Wang, Jiarui Jin, Runhao Fu, Zexuan Yan 외 arxiv

Research agents have recently achieved significant progress in information seeking and synthesis across heterogeneous textual and visual sources. In this paper, we introduce MuSEAgent, a multimodal reasoning agent that e…

Multimodal Reasoning

Sleep-time Compute: Beyond Inference Scaling at Test-time

2025-04-17 · Kevin Lin, Charlie Snell, Yu Wang, Charles Packer 외

Scaling test-time compute has emerged as a key ingredient for enabling large language models (LLMs) to solve difficult problems, but comes with high latency and inference cost. We introduce sleep-time compute, which allo…

FluxEDA: A Unified Execution Infrastructure for Stateful Agentic EDA

2026-03-26 · Zhengrui Chen, Zixuan Song, Yu Li, Qi Sun 외 arxiv

Large language models and autonomous agents are increasingly explored for EDA automation, but many existing integrations still rely on script-level or request-level interactions, which makes it difficult to preserve tool…

Agent-World: Scaling Real-World Environment Synthesis for Evolving General Agent Intelligence

2026-04-20 · Guanting Dong, Junting Lu, Junjie Huang, Wanjun Zhong 외 arxiv

Large language models are increasingly expected to serve as general-purpose agents that interact with external, stateful tool environments. The Model Context Protocol (MCP) and broader agent skills offer a unified interf…

Reinforcement Learning