Contextual Integrity in LLMs via Reasoning and Reinforcement Learning
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909977946357760 |
|---|---|
| author | Lan, Guangchen Inan, Huseyin A. Abdelnabi, Sahar Kulkarni, Janardhan Wutschitz, Lukas Shokri, Reza Brinton, Christopher G. Sim, Robert |
| author_facet | Lan, Guangchen Inan, Huseyin A. Abdelnabi, Sahar Kulkarni, Janardhan Wutschitz, Lukas Shokri, Reza Brinton, Christopher G. Sim, Robert |
| contents | As the era of autonomous agents making decisions on behalf of users unfolds, ensuring contextual integrity (CI) -- what is the appropriate information to share while carrying out a certain task -- becomes a central question to the field. We posit that CI demands a form of reasoning where the agent needs to reason about the context in which it is operating. To test this, we first prompt LLMs to reason explicitly about CI when deciding what information to disclose. We then extend this approach by developing a reinforcement learning (RL) framework that further instills in models the reasoning necessary to achieve CI. Using a synthetic, automatically created, dataset of only $\sim700$ examples but with diverse contexts and information disclosure norms, we show that our method substantially reduces inappropriate information disclosure while maintaining task performance across multiple model sizes and families. Importantly, improvements transfer from this synthetic dataset to established CI benchmarks such as PrivacyLens that has human annotations and evaluates privacy leakage of AI assistants in actions and tool calls. Our code is available at: https://github.com/EricGLan/CI-RL |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_04245 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Contextual Integrity in LLMs via Reasoning and Reinforcement Learning Lan, Guangchen Inan, Huseyin A. Abdelnabi, Sahar Kulkarni, Janardhan Wutschitz, Lukas Shokri, Reza Brinton, Christopher G. Sim, Robert Artificial Intelligence Computation and Language Machine Learning I.2.6; I.2.7 As the era of autonomous agents making decisions on behalf of users unfolds, ensuring contextual integrity (CI) -- what is the appropriate information to share while carrying out a certain task -- becomes a central question to the field. We posit that CI demands a form of reasoning where the agent needs to reason about the context in which it is operating. To test this, we first prompt LLMs to reason explicitly about CI when deciding what information to disclose. We then extend this approach by developing a reinforcement learning (RL) framework that further instills in models the reasoning necessary to achieve CI. Using a synthetic, automatically created, dataset of only $\sim700$ examples but with diverse contexts and information disclosure norms, we show that our method substantially reduces inappropriate information disclosure while maintaining task performance across multiple model sizes and families. Importantly, improvements transfer from this synthetic dataset to established CI benchmarks such as PrivacyLens that has human annotations and evaluates privacy leakage of AI assistants in actions and tool calls. Our code is available at: https://github.com/EricGLan/CI-RL |
| title | Contextual Integrity in LLMs via Reasoning and Reinforcement Learning |
| topic | Artificial Intelligence Computation and Language Machine Learning I.2.6; I.2.7 |
| url | https://arxiv.org/abs/2506.04245 |