Beyond Local vs. External: A Game-Theoretic Framework for Trustworthy Knowledge Acquisition

2026-04-25

AI Summary

This work addresses the challenge of leveraging powerful cloud-based large language models to acquire high-quality knowledge while safeguarding users’ sensitive intents. The authors propose GTKA, a novel framework that formalizes the privacy–utility trade-off as a game-theoretic problem, integrating a privacy-aware sub-query generator, an adversarial reconstruction attacker, and a trusted local integrator in a协同 optimization process. GTKA employs an alternating adversarial training mechanism to dynamically refine query strategies, enabling efficient external knowledge integration within provable privacy boundaries. Evaluated on sensitive-domain benchmarks—including biomedical and legal tasks—GTKA substantially reduces intent leakage risk without compromising response quality, achieving performance on par with state-of-the-art baselines.

Technology Category

• Machine Learning: Privacy
• Data Mining & Knowledge Management: Intelligent Query Processing
• Natural Language Processing: Safety and Robustness

Application Category

• Search and Retrieval-Augmented AI: Large language models for search
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Cloud-hosted Large Language Models (LLMs) offer unmatched reasoning capabilities and dynamic knowledge, yet submitting raw queries to these external services risks exposing sensitive user intent. Conversely, relying exclusively on trusted local models preserves privacy but often compromises answer quality due to limited parameter scale and knowledge. To resolve this dilemma, we propose Game-theoretic Trustworthy Knowledge Acquisition (GTKA), a framework that formulates the trade-off between knowledge utility and privacy as a strategic game. GTKA consists of three components: (i) a privacy-aware sub-query generator that decomposes sensitive intent into generalized, low-risk fragments; (ii) an adversarial reconstruction attacker that attempts to infer the original query from these fragments, providing adaptive leakage signals; and (iii) a trusted local integrator that synthesizes external responses within a secure boundary. By training the generator and attacker in an alternating adversarial manner, GTKA optimizes the sub-query generation policy to maximize knowledge acquisition accuracy while minimizing the reconstructability of the original sensitive intent. To validate our approach, we construct two sensitive-domain benchmarks in the biomedical and legal fields. Extensive experiments demonstrate that GTKA significantly reduces intent leakage compared to state-of-the-art baselines while maintaining high-fidelity answer quality.

Problem

• privacy
• knowledge acquisition
• large language models
• trustworthy AI
• intent leakage

Innovation

• game-theoretic framework
• privacy-preserving query decomposition
• adversarial reconstruction attacker
• trusted local integration
• knowledge acquisition