# PrivAct: Internalizing Contextual Privacy Preservation via Multi-Agent Preference Training

arXiv: https://arxiv.org/abs/2602.13840 · HTML: https://mlsanity.com/paper/2602.13840/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Placed in Agent threats, guardrails and governance, one of 16 directions in LLM agents and tool use.
- Builds on 38 of its 47 references that are in the map.
- First posted on arXiv 14 Feb 2026.

## Placed in

- LLM agents and tool use / [Agent threats, guardrails and governance](https://mlsanity.com/agents/agent-threats-guardrails-and-governance/)

## Builds on (most cited first, 30 of 38)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [BeaverTails: Towards Improved Safety Alignment of LLM via a Human-Preference Dataset](https://mlsanity.com/paper/2307.04657/) (2023)
- [DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelines](https://mlsanity.com/paper/2310.03714/) (2023)
- [Analyzing Leakage of Personally Identifiable Information in Language Models](https://mlsanity.com/paper/2302.00539/) (2023)
- [Step-DPO: Step-wise Preference Optimization for Long-chain Reasoning of LLMs](https://mlsanity.com/paper/2406.18629/) (2024)
- [Can LLMs Keep a Secret? Testing Privacy Implications of Language Models via Contextual Integrity Theory](https://mlsanity.com/paper/2310.17884/) (2023)
- [Chain of Agents: Large Language Models Collaborating on Long-Context Tasks](https://mlsanity.com/paper/2406.02818/) (2024)
- [PrivacyLens: Evaluating Privacy Norm Awareness of Language Models in Action](https://mlsanity.com/paper/2409.00138/) (2024)
- [ProPILE: Probing Privacy Leakage in Large Language Models](https://mlsanity.com/paper/2307.01881/) (2023)
- [Multiagent Finetuning: Self Improvement with Diverse Reasoning Chains](https://mlsanity.com/paper/2501.05707/) (2025)
- [The Hidden Risks of Large Reasoning Models: A Safety Assessment of R1](https://mlsanity.com/paper/2502.12659/) (2025)
- [MAPoRL: Multi-Agent Post-Co-Training for Collaborative Large Language Models with Reinforcement Learning](https://mlsanity.com/paper/2502.18439/) (2025)
- [AgentDAM: Privacy Leakage Evaluation for Autonomous Web Agents](https://mlsanity.com/paper/2503.09780/) (2025)
- [MALT: Improving Reasoning with Multi-Agent LLM Training](https://mlsanity.com/paper/2412.01928/) (2024)
- [In-the-Flow Agentic System Optimization for Effective Planning and Tool Use](https://mlsanity.com/paper/2510.05592/) (2025)
- [Multi-Agent Evolve: LLM Self-Improve through Co-evolution](https://mlsanity.com/paper/2510.23595/) (2025)
- [G-Safeguard: A Topology-Guided Security Lens and Treatment on LLM-based Multi-agent Systems](https://mlsanity.com/paper/2502.11127/) (2025)
- [Learning Adaptive Parallel Reasoning with Language Models](https://mlsanity.com/paper/2504.15466/) (2025)
- [SiriuS: Self-improving Multi-agent Systems via Bootstrapped Reasoning](https://mlsanity.com/paper/2502.04780/) (2025)
- [Parametric Retrieval Augmented Generation](https://mlsanity.com/paper/2501.15915/) (2025)
- [Privacy in Action: Towards Realistic Privacy Mitigation and Evaluation for LLM-Powered Agents](https://mlsanity.com/paper/2509.17488/) (2025)
- [Step-Controlled DPO: Leveraging Stepwise Error for Enhanced Mathematical Reasoning](https://mlsanity.com/paper/2407.00782/) (2024)
- [Leaky Thoughts: Large Reasoning Models Are Not Private Thinkers](https://mlsanity.com/paper/2506.15674/) (2025)
- [Searching for Privacy Risks in LLM Agents via Simulation](https://mlsanity.com/paper/2508.10880/) (2025)
- [The Sum Leaks More Than Its Parts: Compositional Privacy Risks and Mitigations in Multi-Agent Collaboration](https://mlsanity.com/paper/2509.14284/) (2025)
- [Talk Structurally, Act Hierarchically: A Collaborative Framework for LLM Multi-Agent Systems](https://mlsanity.com/paper/2502.11098/) (2025)
- [1-2-3 Check: Enhancing Contextual Privacy in LLM via Multi-Agent Reasoning](https://mlsanity.com/paper/2508.07667/) (2025)
- [LeakAgent: RL-based Red-teaming Agent for LLM Privacy Leakage](https://mlsanity.com/paper/2412.05734/) (2024)

## Built on by (most cited first, 7 of 7)

- [LCGuard: Latent Communication Guard for Safe KV Sharing in Multi-Agent Systems](https://mlsanity.com/paper/2605.22786/) (2026)
- [PrivacyAlign: Contextual Privacy Alignment for LLM Agents](https://mlsanity.com/paper/2606.21710/) (2026)
- [What Intermediate Layers Know: Detecting Jailbreaks from Entropy Dynamics](https://mlsanity.com/paper/2606.25182/) (2026)
- [Reinforcing privacy reasoning in LLMs via normative simulacra from fiction](https://mlsanity.com/paper/2604.20904/) (2026)
- [Confuse the Model, Control the Flow: Understanding and Mitigating Privacy Leakage from LLM Agents with Information Flow Control](https://mlsanity.com/paper/2609.14003/) (2026)
- [Federated Agent Optimization](https://mlsanity.com/paper/2610.01195/) (2026)
- [Loose Lips: Understanding Privacy Leakage in Agentic Dialogue](https://mlsanity.com/paper/2610.14071/) (2026)

## Abstract (arXiv)

Large language model (LLM) agents are increasingly deployed in personalized tasks involving sensitive, context-dependent information, where privacy violations may arise in agents' action due to the implicitness of contextual privacy. Existing approaches rely on external, inference-time interventions which are brittle, scenario-specific, and may expand the privacy attack surface. We propose PrivAct, a contextual privacy-aware multi-agent learning framework that internalizes contextual privacy preservation directly into models' generation behavior for privacy-compliant agentic actions. By embedding privacy preferences into each agent, PrivAct enhances system-wide contextual integrity while achieving a more favorable privacy-helpfulness tradeoff. Experiments across multiple LLM backbones and benchmarks demonstrate consistent improvements in contextual privacy preservation, reducing leakage rates by up to 12.32% while maintaining comparable helpfulness, as well as zero-shot generalization and robustness across diverse multi-agent topologies. Code is available at https://github.com/chengyh23/PrivAct.