arXiv cs.AI· Zihou Ren, Byungchul Tak, Long Wang·· 14 小时前AI 评分40
REALM：为 LLM 智能体识别最小任务资产
REALM: Identifying Minimal Task Assets for LLM Agents
AI 导读
REALM 通过将任务所需资产表示为依赖图节点，把资产发现与使用授权分离，从而为 LLM 智能体落实最小权限。在 AgentDyn 的 60 个干净任务和 560 个注入案例上，以 qwen3.5-flash 为基座模型，任务完成率为 40.0% 和 36.4%，ASR 从 16.6% 降至 0.2%。
正文 · 原文
Abstract:Large language model (LLM) agents often receive access to more files, accounts, and services than a task requires. Prompt injection turns this excess access into an opportunity to redirect execution toward unrelated resources. Applying least privilege therefore requires identifying the assets needed by a task, including assets whose identities emerge only during execution. We present REALM (Required Execution Assets Linked to a Mission), an approach to identifying and maintaining a task's required assets and their permitted uses. The approach represents required assets as Dependency Graph nodes, resolves prerequisite relationships, and admits newly discovered assets through source-scoped selection. It separates discovering an asset from authorizing its use, while preserving the agent's freedom to choose its computation. The REALM Builder maintains the selected grants and projects Asset Files; the Partial API Manager exposes resource-bound Partial APIs to the Agent inside a Task Container. Across 60 clean tasks and 560 injected cases in AgentDyn, it completes 40.0% and 36.4% of tasks and decreases ASR from 16.6% to 0.2% with qwen3.5-flash as the base model.
| Subjects: | Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Systems and Control (eess.SY) |
| Cite as: | arXiv:2610.39155 [cs.CR] |
| (or arXiv:2610.39155v1 [cs.CR] for this version) | |
| https://doi.org/10.48550/arXiv.2610.39155 arXiv-issued DOI via DataCite |
Submission history
From: Zihou Ren [view email]
[v1]
Thu, 29 Oct 2026 01:40:42 UTC (171 KB)
来源：arXiv cs.AI · arxiv.org