← 研踪 PaperTrailarXiv:2611.20045 · v1
Security of Deep Research Agents: Attack Surface, Defenses, Benchmarks, and Agenda Seeding against Research Planning
首次提交 2026-11-15 22:44
版本更新 2026-11-15 22:44
雷达观测 2026-11-17 12:06
摘要
Deep research and agentic search agents autonomously retrieve content from the open web and generate reports with citations, and their outputs are often taken directly as research conclusions. This paper systematizes security research in this area along four stages, namely retrieval, planning, evidence integration, and report generation. We present a unified threat model, a taxonomy of attacks and defenses, and a review of evaluation benchmarks, and we align the area dimension by dimension with retrieval-augmented generation (RAG) poisoning, prompt injection, and generative engine optimization (GEO). Coding 34 primary studies shows that only 26.5% of them involve the planning stage, and none performs an end-to-end poisoning evaluation on commercial products. To address this gap, we propose the agenda seeding attack, in which the attacker writes no instructions and only introduces a fictitious evaluation metric. Across 3,520 sandboxed research runs under the same budget, GEO and prompt injection are nearly ineffective, whereas agenda seeding makes the attacker's target the top pick in 16% to 28% of reports. Injection detection and query anchoring fail against it, verification prompting hurts utility, and our proposed provenance tagging preserves utility but can be bypassed by multi-site bait. Finally, we propose eight research directions.
作者备注:31 pages, 5 figures
观测到的版本
相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。
v1 · 2026-11-15 22:44 · 最新观测版本
Security of Deep Research Agents: Attack Surface, Defenses, Benchmarks, and Agenda Seeding against Research Planning
Deep research and agentic search agents autonomously retrieve content from the open web and generate reports with citations, and their outputs are often taken directly as research conclusions. This paper systematizes security research in this area along four stages, namely retrieval, planning, evidence integration, and report generation. We present a unified threat model, a taxonomy of attacks and defenses, and a review of evaluation benchmarks, and we align the area dimension by dimension with retrieval-augmented generation (RAG) poisoning, prompt injection, and generative engine optimization (GEO). Coding 34 primary studies shows that only 26.5% of them involve the planning stage, and none performs an end-to-end poisoning evaluation on commercial products. To address this gap, we propose the agenda seeding attack, in which the attacker writes no instructions and only introduces a fictitious evaluation metric. Across 3,520 sandboxed research runs under the same budget, GEO and prompt injection are nearly ineffective, whereas agenda seeding makes the attacker's target the top pick in 16% to 28% of reports. Injection detection and query anchoring fail against it, verification prompting hurts utility, and our proposed provenance tagging preserves utility but can be bypassed by multi-site bait. Finally, we propose eight research directions.
查看此版本解读 →查看此版本原文 ↗