MosaicLeaks:Privacy Risks in Querying-in-the-Open for Deep Research Agents

• URL: http://arxiv.org/abs/2605.30727v1
• Date: Fri, 29 May 2026 01:44:35 GMT
• Title: MosaicLeaks:Privacy Risks in Querying-in-the-Open for Deep Research Agents
• Abstract summary: Deep research agents increasingly combine private local documents with external tools like web retrieval, creating a privacy risk.<n>We introduce MosaicLeaks, a benchmark of 1,001 deep research tasks that chain private enterprise documents and a public web corpus.<n>We find that models across families and sizes frequently leak at all three levels, that zero-shot privacy prompting reduces but does not eliminate leakage.
• Score: 76.60965179634664
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Deep research agents increasingly combine private local documents with external tools like web retrieval, creating a privacy risk: an agent's external queries may leak sensitive information from its local context. This risk is amplified by the mosaic effect, where individual queries may appear harmless but become revealing in aggregate. We introduce MosaicLeaks, a benchmark of 1,001 multi-hop deep research tasks that chain private enterprise documents and a public web corpus, forcing agents to make external queries that depend on local information. We evaluate leakage with an adversary LLM that observes only the agent's external queries and attempts to infer private information at three levels: the agent's research intent, answers to specific private questions and verifiable claims about the enterprise documents. We find that models across families and sizes frequently leak at all three levels, that zero-shot privacy prompting reduces but does not eliminate leakage and that reinforcement learning for task performance alone worsens leakage. To address this, we propose Privacy-Aware Deep Research (PA-DR), an RL framework that combines situational rewards for task success with a learned privacy classifier to provide dense credit assignment over both per-query and mosaic-level leakage. Training Qwen3-4B-Instruct with PA-DR improves accuracy from 48.7% to 58.7% and reduces answer and full-information leakage from 34.0% to 9.9%.

Related papers

• PrivacyPeek: Auditing What LLM-Based Agents Acquire, Not Just What They Say [48.84133320567554]
  LLM-based agents are rapidly advancing, autonomously invoking external tools to complete tasks for users.<n> Existing privacy benchmarks audit what the agent's response or outgoing actions disclose, but overlook the acquisition stage where data first enters the agent's context.<n>We introduce emphPrivacyPeek, a benchmark for evaluating acquisition-stage privacy leakage of LLM-based agents.
  arXiv  Detail & Related papers  (2026-05-29T04:55:12Z)
• Profiling for Pennies: Unveiling the Privacy Iceberg of LLM Agents [52.02752011119451]
  We investigate privacy perception in the real world through the lens of human awareness.<n>We propose the PrivacyIceberg, which categorizes real-world human privacy risks into three tiers.
  arXiv  Detail & Related papers  (2026-05-07T13:21:44Z)
• SafeSearch: Automated Red-Teaming for the Safety of LLM-Based Search Agents [63.70653857721785]
  We conduct two in-the-wild experiments to demonstrate the prevalence of low-quality search results and their potential to misguide agent behaviors.<n>To counter this threat, we introduce an automated red-teaming framework that is systematic, scalable, and cost-efficient.
  arXiv  Detail & Related papers  (2025-09-28T07:05:17Z)
• 1-2-3 Check: Enhancing Contextual Privacy in LLM via Multi-Agent Reasoning [18.751008976082655]
  We introduce a multi-agent framework that decomposes privacy reasoning into specialized subtasks (extraction, classification)<n>We conduct a systematic ablation over information-flow topologies, revealing when and why upstream detection mistakes cascade into downstream leakage.
  arXiv  Detail & Related papers  (2025-08-11T06:34:09Z)
• MAGPIE: A dataset for Multi-AGent contextual PrIvacy Evaluation [54.410825977390274]
  Existing benchmarks to evaluate contextual privacy in LLM-agents primarily assess single-turn, low-complexity tasks.<n>We first present a benchmark - MAGPIE comprising 158 real-life high-stakes scenarios across 15 domains.<n>We then evaluate the current state-of-the-art LLMs on their understanding of contextually private data and their ability to collaborate without violating user privacy.
  arXiv  Detail & Related papers  (2025-06-25T18:04:25Z)
• Privacy Leakage Overshadowed by Views of AI: A Study on Human Oversight of Privacy in Language Model Agent [3.89966779727297]
  Language model (LM) agents that act on users' behalf for personal tasks can boost productivity, but are also susceptible to unintended privacy leakage risks.<n>We present the first study on people's capacity to oversee the privacy implications of the LM agents.
  arXiv  Detail & Related papers  (2024-11-02T19:15:42Z)
• PrivacyLens: Evaluating Privacy Norm Awareness of Language Models in Action [54.11479432110771]
  PrivacyLens is a novel framework designed to extend privacy-sensitive seeds into expressive vignettes and further into agent trajectories.<n>We instantiate PrivacyLens with a collection of privacy norms grounded in privacy literature and crowdsourced seeds.<n>State-of-the-art LMs, like GPT-4 and Llama-3-70B, leak sensitive information in 25.68% and 38.69% of cases, even when prompted with privacy-enhancing instructions.
  arXiv  Detail & Related papers  (2024-08-29T17:58:38Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.