ScholarPulse 日报 2026-07-28

2026-07-28 学术简报：2 篇。APP A框架通过引擎管理的上下文分支和前瞻性授权机制，有效缓解LLM代理处理混合机密数据时因传统taint tracking导致的效用损失问题。

今日速览

序号	标题	来源	日期	主题	推荐等级
1	Agentic Permissions Policy Algebra for Taint Confinement in LLM Agents	arXiv	2026-07-27	AI-Agent	高
2	SIREN: Towards End-to-End Extreme-Weather Early Warning with Experience-Grounded LLM Agents	arXiv	2026-07-27	AI-Agent	高

重点论文与技术动态

1. Agentic Permissions Policy Algebra for Taint Confinement in LLM Agents

来源：arXiv
日期：2026-07-27
作者/机构：Arseny Kravchenko, Vadim Liventsev, Innokentii Konstantinov, Ildar Iskhakov, Matvey Kukuy
主题标签：AI-Agent, arXiv
推荐等级：高
分类：cs.CR, cs.AI

一句话结论

APP A框架通过引擎管理的上下文分支和前瞻性授权机制，有效缓解LLM代理处理混合机密数据时因传统taint tracking导致的效用损失问题。

核心内容

传统taint tracking永久污染代理上下文，严重限制下游功能。
APP A在数据获取前前瞻性评估标签下降和缺失前提，生成授权或接受计划，避免污染主上下文。
评估显示APP A将数据泄露攻击成功率从31%-50%降至0%-7%，并在三个模型中恢复了taint tracking损失的效用。

方法与数据

方法：基于安全标签的双单群模型，通过标签种子子轨迹实现局部标签下降吸收。
数据：在四个模型的多轮工具链基准测试上验证。

价值判断

值得关注：显著降低prompt injection攻击风险，提升LLM代理安全性。
可复用点：上下文分支和前瞻性评估机制可推广至其他动态信息流控制框架。
局限/待核查：摘要未明确。

摘要

Autonomous LLM agents processing mixed-confidentiality data face severe security risks from prompt injection attacks and reasoning errors. While dynamic Information Flow Control (IFC) provides structural security guarantees, traditional taint tracking permanently taints an agent's context upon reading unvetted data, severely restricting downstream utility. We present APPA (Agentic Permissions Policy Algebra), an IFC framework that resolves this usability bottleneck through engine-managed context branching and prospective acquisition enforcement. Before data acquisition occurs, APPA prospectively evaluates label descents and missing prerequisites, generating actionable remedy plans (Authorize, Accept). To inspect unvetted data without polluting the primary context, a label-seeded child trajectory is spawned, absorbing label descent locally and allowing a trusted sanitizer to return a bounded derivative to the unchanged parent. Governed by a two-monoid model over security labels and shared event logs, we formally prove parent label preservation and merge confinement. Finally, we evaluate APPA on a multi-turn tool-chaining benchmark across four models: it suppresses exfiltration (31%-50% down to 0%-7% attack success), and on three of the four, branching recovers a substantial share of the utility that taint tracking alone forfeits.

2. SIREN: Towards End-to-End Extreme-Weather Early Warning with Experience-Grounded LLM Agents

来源：arXiv
日期：2026-07-27
作者/机构：Hang Ni, Weijia Zhang, Fan Liu, Mengqian Lu, Hao Liu
主题标签：AI-Agent, arXiv
推荐等级：高
分类：cs.AI

一句话结论

SIREN框架通过经验驱动的LLM代理实现了端到端极端天气预警自动化，在个体预警步骤和端到端链上均显著优于现有天气代理基线。

核心内容

现有天气代理研究仅聚焦孤立科学任务，忽视操作性预警所需的多步骤依赖流程，导致能力缺口。
开发SIREN-Bench基准，包含600个问答实例覆盖19个任务及四个个体预警步骤与端到端预警链。
SIREN框架整合异构天气证据与工具的代理执行环境，并通过检索、技能蒸馏和预测建模利用历史案例优化预警。

方法与数据

SIREN-Bench基准用于评估代理能力，包含600个问答实例。
摘要未明确具体数据来源或实验细节。

价值判断

值得关注：解决专家中心预警工作流成本高、难扩展的痛点，提升极端天气风险应对效率。
可复用点：经验 grounded 的LLM代理设计模式，可迁移至其他需历史案例的自动化任务领域。
局限/待核查：摘要未明确提及框架局限性或需验证的方面。

（字数：298）

摘要

Early warning of extreme weather is essential for mitigating the societal, economic, and environmental risks posed by hazardous weather events. However, expert-centered warning workflows are costly, labor-intensive, and difficult to scale throughout the warning-to-action process. Although recent advances in Large Language Model (LLM) agents have enabled the automation of weather-related tasks, existing studies remain centered on isolated scientific tasks and overlook the chain of interdependent processes required for operational extreme-weather early warning. To bridge this gap, this study investigates automated end-to-end extreme-weather early warning through LLM agents. We first develop SIREN-Bench, a comprehensive benchmark comprising 600 question-answer instances across 19 tasks, and covering four individual warning procedures and an end-to-end warning chain. Evaluation on SIREN-Bench reveals substantial capability gaps in existing weather agent frameworks. This motivates us to develop SIREN, an experience-grounded agent framework inspired by experts' use of historical cases, which combines an agentic execution environment integrating heterogeneous weather evidence and tools with a family of agent harnesses that exploit historical cases through retrieval, skill distillation, and predictive modeling. Extensive experiments demonstrate that SIREN outperforms weather-agent baselines on both individual warning procedures and end-to-end warning chains.