arXiv AI
Evaluating Deep-Search Agents under Hierarchical Web Evidence Poisoning
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
arXiv AI
Aug 6
Breadcrumbing Search Agents
arXiv:2608. 04565v1 Announce Type: cross Abstract: LLM-based search agents are widely used for information-seeking tasks, but their reliance on external tool returns introduces a critical security risk: web content retrieved during execution is untrusted, exposing agents to prompt injection and goal hijacking.
arXiv AI
6d ago
Breadcrumbing Search Agents: Per-Turn Scheming Over Long-Horizon Trajectories
The paper investigates how large language model (LLM) search agents can be manipulated through repeated, coordinated search-result injections, a threat that extends beyond single-page poisoning. It introduces Authority‑Chain Hijack (ACH), a strategy that chains manipulated evidence across multiple sources, achieving high attack success rates on the SafeSearch benchmark. The authors also present Trace‑Guided Strategy Evolution (TGSE), an automated method that refines attacker strategies using execution traces, further improving success rates.
arXiv AI
Jun 12
One Polluted Page Is Enough: Evaluating Web Content Pollution in Generative Recommenders
arXiv:2606. 13610v1 Announce Type: cross Abstract: Search-augmented LLMs increasingly mediate everyday consumer recommendations by retrieving live web content.
arXiv AI
Jun 30
SafeGEO: Understanding Generative Engine Optimization Risks in Recommendation Agents
arXiv:2606. 28356v1 Announce Type: cross Abstract: Generative Engine Optimization (GEO) lets content owners rewrite web content to increase their visibility in generative systems.
arXiv AI
Aug 25
One Polluted Page Is Enough: Evaluating Web Content Pollution in LLM Recommenders
The paper introduces FORGE, a benchmark that rewrites real product pages into fake ones to test how often search‑augmented large language models (LLMs) recommend these polluted items. Across 12 commercial and open‑weight LLMs, a single polluted page can lead to up to 27% of recommendations being fake, rising to 73.8% when the top‑3 replacements are used. The study finds that reasoning does not help and existing defenses—skepticism prompts, consensus filters, and credibility re‑ranking—are largely ineffective.
arXiv AI
Jul 7
FORGE: Research-Trajectory Hijacking Attacks on Deep Research Agents
arXiv:2607. 04718v1 Announce Type: new Abstract: Deep research agents decompose open-ended queries into subtasks, retrieve web evidence over multiple rounds, and synthesize long-form reports.