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

arXiv: https://arxiv.org/abs/2604.23413 · HTML: https://mlsanity.com/paper/2604.23413/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning.
- Builds on 15 of its 39 references that are in the map.
- First posted on arXiv 25 Apr 2026.

## Placed in

- LLM reasoning / [Reasoning for agents, planning and code](https://mlsanity.com/lr/reasoning-for-agents-planning-and-code/)

## Builds on (most cited first, 15 of 15)

- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Explaining and Harnessing Adversarial Examples](https://mlsanity.com/paper/1412.6572/) (2014)
- [Towards Deep Learning Models Resistant to Adversarial Attacks](https://mlsanity.com/paper/1706.06083/) (2017)
- [Deep Learning with Differential Privacy](https://mlsanity.com/paper/1607.00133/) (2016)
- [Threats to Federated Learning: A Survey](https://mlsanity.com/paper/2003.02133/) (2020)
- [PPFL: Privacy-preserving Federated Learning with Trusted Execution Environments](https://mlsanity.com/paper/2104.14380/) (2021)
- [Locally Differentially Private Document Generation Using Zero Shot Prompting](https://mlsanity.com/paper/2310.16111/) (2023)
- [Protecting User Privacy in Remote Conversational Systems: A Privacy-Preserving framework based on text sanitization](https://mlsanity.com/paper/2306.08223/) (2023)
- [End-to-End Learning and Intervention in Games](https://mlsanity.com/paper/2010.13834/) (2020)
- [Meta Stackelberg Game: Robust Federated Learning against Adaptive and Mixed Poisoning Attacks](https://mlsanity.com/paper/2410.17431/) (2024)
- [Privacy-Preserving Language Model Inference with Instance Obfuscation](https://mlsanity.com/paper/2402.08227/) (2024)
- [Automating Knowledge Acquisition for Content-Centric Cognitive Agents Using LLMs](https://mlsanity.com/paper/2312.16378/) (2023)
- [Machine Learning Techniques for Stackelberg Security Games: a Survey](https://mlsanity.com/paper/1609.09341/) (2016)
- [Learning from Response not Preference: A Stackelberg Approach for LLM Detoxification using Non-parallel Data](https://mlsanity.com/paper/2410.20298/) (2024)

## Abstract (arXiv)

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.