# Need to Know: Contextual-Integrity-Grounded Query Rewriting for Privacy-Conscious LLM Delegation

Source: [Need to Know: Contextual-Integrity-Grounded Query Rewriting for Privacy-Conscious LLM Delegation](<https://www.sota2.com/research/paper/6a2106dec10484bac23f0508>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-06-05T05:00:04.417Z

Authors: Xinyue Huang, Xiaochun Cao, Wenyuan Yang

Published: 2026-06-02T14:28:28.000Z

## Abstract

As LLMs become increasingly woven into everyday workflows, user queries sent to cloud hosted LLMs routinely mix task-essential content with task non-essential sensitive disclosures, yet type based PII redaction is context agnostic and may raise two issues: over disclosing untyped sensitive context and over removing answer bearing spans. We recast privacy preserving query rewriting under Contextual Integrity: a span should be forwarded only if it is necessary for the task. We introduce DelegateCI-Bench, the first task based Contextual Integrity benchmark for privacy-conscious delegation, comprising 3,167 samples that combine high quality synthetic data spanning 11 tasks and 20 task types, WildChat based real user queries, and a medical challenge set with dense sensitive information. Building on this benchmark, we propose a CI-guided reinforcement learning framework that converts essential and non-essential sensitive spans into verifiable optimization signals, and train a query rewriter to preserve task critical information while suppressing unnecessary sensitive disclosure. Experiments show that our learned rewriter achieves the best privacy-utility tradeoff, achieving up to +10.1 average utility over on-device baselines.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Privacy-preserving Query Rewriting on DelegateCI-Bench Medical](<https://www.sota2.com/research/sota/privacy-preserving-query-rewriting-on-delegateci-bench-medical.md>): Privacy-preserving Query Rewriting; DelegateCI-Bench Medical; Utilization: 66.7.

- [Privacy-preserving Query Rewriting on DelegateCI-Bench General](<https://www.sota2.com/research/sota/privacy-preserving-query-rewriting-on-delegateci-bench-general.md>): Privacy-preserving Query Rewriting; DelegateCI-Bench General; Utilization: 95.3.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2606.04067>)

[PDF](<https://arxiv.org/pdf/2606.04067.pdf>)