# PrivacyAlign: Contextual Privacy Alignment for LLM Agents

arXiv: https://arxiv.org/abs/2606.21710 · HTML: https://mlsanity.com/paper/2606.21710/

## Key facts

- Cited by 2 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 36 of its 45 references that are in the map.
- First posted on arXiv 19 Jun 2026.

## Placed in

- LLM agents and tool use / [Agent threats, guardrails and governance](https://mlsanity.com/agents/agent-threats-guardrails-and-governance/)
- LLM alignment / RLHF / preference optimisation / [Human values and AI feedback alignment](https://mlsanity.com/alignment/human-values-and-ai-feedback-alignment/)

## Builds on (most cited first, 30 of 36)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [Toolformer: Language Models Can Teach Themselves to Use Tools](https://mlsanity.com/paper/2302.04761/) (2023)
- [Extracting Training Data from Large Language Models](https://mlsanity.com/paper/2012.07805/) (2020)
- [RLAIF vs. RLHF: Scaling Reinforcement Learning from Human Feedback with AI Feedback](https://mlsanity.com/paper/2309.00267/) (2023)
- [Rubrics as Rewards: Reinforcement Learning Beyond Verifiable Domains](https://mlsanity.com/paper/2507.17746/) (2025)
- [Judging the Judges: A Systematic Study of Position Bias in LLM-as-a-Judge](https://mlsanity.com/paper/2406.07791/) (2024)
- [Can LLMs Keep a Secret? Testing Privacy Implications of Language Models via Contextual Integrity Theory](https://mlsanity.com/paper/2310.17884/) (2023)
- [PrivacyLens: Evaluating Privacy Norm Awareness of Language Models in Action](https://mlsanity.com/paper/2409.00138/) (2024)
- [Soft Adaptive Policy Optimization](https://mlsanity.com/paper/2511.20347/) (2025)
- [AirGapAgent: Protecting Privacy-Conscious Conversational Agents](https://mlsanity.com/paper/2405.05175/) (2024)
- [AgentDAM: Privacy Leakage Evaluation for Autonomous Web Agents](https://mlsanity.com/paper/2503.09780/) (2025)
- [PAPILLON: Privacy Preservation from Internet-based and Local Language Model Ensembles](https://mlsanity.com/paper/2410.17127/) (2024)
- [Be like a Goldfish, Don't Memorize! Mitigating Memorization in Generative LLMs](https://mlsanity.com/paper/2406.10209/) (2024)
- [Operationalizing Contextual Integrity in Privacy-Conscious Assistants](https://mlsanity.com/paper/2408.02373/) (2024)
- [CI-Bench: Benchmarking Contextual Integrity of AI Assistants on Synthetic Data](https://mlsanity.com/paper/2409.13903/) (2024)
- [Privacy in Action: Towards Realistic Privacy Mitigation and Evaluation for LLM-Powered Agents](https://mlsanity.com/paper/2509.17488/) (2025)
- [Contextual Integrity in LLMs via Reasoning and Reinforcement Learning](https://mlsanity.com/paper/2506.04245/) (2025)
- [MAGPIE: A benchmark for Multi-AGent contextual PrIvacy Evaluation](https://mlsanity.com/paper/2510.15186/) (2025)
- [Leaky Thoughts: Large Reasoning Models Are Not Private Thinkers](https://mlsanity.com/paper/2506.15674/) (2025)
- [CIMemories: A Compositional Benchmark for Contextual Integrity of Persistent Memory in LLMs](https://mlsanity.com/paper/2511.14937/) (2025)
- [Benchmarking LLM Faithfulness in RAG with Evolving Leaderboards](https://mlsanity.com/paper/2505.04847/) (2025)
- [The Sum Leaks More Than Its Parts: Compositional Privacy Risks and Mitigations in Multi-Agent Collaboration](https://mlsanity.com/paper/2509.14284/) (2025)
- [1-2-3 Check: Enhancing Contextual Privacy in LLM via Multi-Agent Reasoning](https://mlsanity.com/paper/2508.07667/) (2025)
- [Position: Privacy Is Not Just Memorization!](https://mlsanity.com/paper/2510.01645/) (2025)
- [CI-Work: Benchmarking Contextual Integrity in Enterprise LLM Agents](https://mlsanity.com/paper/2604.21308/) (2026)

## Built on by (most cited first, 2 of 2)

- ["Nothing to See Here'': Unintended Disclosure through Revision Traces of LLM Deliverables](https://mlsanity.com/paper/2609.35408/) (2026)
- [PrivacySkills: How Privacy Guidance Shapes Source Selection in LLM Agents](https://mlsanity.com/paper/2609.35937/) (2026)

## Abstract (arXiv)

AI agents acting on behalf of users are constantly making decisions, and for users to trust their agents, those decisions must align with what they actually want. Privacy is an important alignment problem for agents: every message, post, or tool call an agent makes is a contextual judgment about what is appropriate to share, with whom, and under which conditions. Because such judgments depend on social expectations and norms, human judgment does not merely label privacy violations but also helps define them. While existing work relies on unreliable proxies for both training and evaluation, we place human judgment at the center of agentic privacy alignment. We introduce PrivacyAlign, a dataset of 1,350 samples with 3,516 detailed annotations from 599 unique annotators across diverse scenarios where current LLMs actually leak, and use it to ground both alignment training and automated evaluation in human privacy norms. Building on these annotations, we first show that conditioning LLM judges on human annotations and explanations for reference responses to the same prompt makes their judgments more reliable. We then introduce annotation-conditioned reward modeling, which uses these annotations to score new responses during RL, and show that small open-weight agents trained with this reward better align with human privacy norms, with strong gains on PrivacyAlign and existing privacy benchmarks for agents.