# PrivGemo: Privacy-Preserving Dual-Tower Graph Retrieval for Empowering LLM Reasoning with Memory Augmentation

arXiv: https://arxiv.org/abs/2601.08739 · HTML: https://mlsanity.com/paper/2601.08739/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Placed in Knowledge- and tool-augmented reasoning, one of 17 directions in LLM reasoning.
- Builds on 16 of its 38 references that are in the map.
- First posted on arXiv 13 Jan 2026.

## Placed in

- LLM reasoning / [Knowledge- and tool-augmented reasoning](https://mlsanity.com/lr/knowledge-and-tool-augmented-reasoning/)

## Builds on (most cited first, 16 of 16)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [Retrieval-Augmented Generation for Large Language Models: A Survey](https://mlsanity.com/paper/2312.10997/) (2023)
- [MemGPT: Towards LLMs as Operating Systems](https://mlsanity.com/paper/2310.08560/) (2023)
- [The Faiss library](https://mlsanity.com/paper/2401.08281/) (2024)
- [StructGPT: A General Framework for Large Language Model to Reason over Structured Data](https://mlsanity.com/paper/2305.09645/) (2023)
- [Reasoning on Graphs: Faithful and Interpretable Large Language Model Reasoning](https://mlsanity.com/paper/2310.01061/) (2023)
- [Knowledge Enhanced Contextual Word Representations](https://mlsanity.com/paper/1909.04164/) (2019)
- [Knowledge-Augmented Language Model Prompting for Zero-Shot Knowledge Graph Question Answering](https://mlsanity.com/paper/2306.04136/) (2023)
- [SCM: Enhancing Large Language Model with Self-Controlled Memory Framework](https://mlsanity.com/paper/2304.13343/) (2023)
- [Generate-on-Graph: Treat LLM as both Agent and KG in Incomplete Knowledge Graph Question Answering](https://mlsanity.com/paper/2404.14741/) (2024)
- [AgentLite: A Lightweight Library for Building and Advancing Task-Oriented LLM Agent System](https://mlsanity.com/paper/2402.15538/) (2024)
- [Privacy-protected Retrieval-Augmented Generation for Knowledge Graph Question Answering](https://mlsanity.com/paper/2508.08785/) (2025)
- [Chase Anonymisation: Privacy-Preserving Knowledge Graphs with Logical Reasoning](https://mlsanity.com/paper/2410.12418/) (2024)

## Built on by (most cited first, 9 of 9)

- [PruneRAG: Confidence-Guided Query Decomposition Trees for Efficient Retrieval-Augmented Generation](https://mlsanity.com/paper/2601.11024/) (2026)
- [An AI Agent Execution Environment to Safeguard User Data](https://mlsanity.com/paper/2604.19657/) (2026)
- [Minim: Privacy-Aware Minimal View for Agents via Trusted Local Sanitization](https://mlsanity.com/paper/2606.13949/) (2026)
- [C2TC: A Training-Free Framework for Efficient Tabular Data Condensation](https://mlsanity.com/paper/2602.21717/) (2026)
- [Anchor-guided Hypergraph Condensation with Dual-level Discrimination](https://mlsanity.com/paper/2605.10001/) (2026)
- [Tackling Multimodal Learning Challenges with Mixture-of-Expert: A Survey](https://mlsanity.com/paper/2605.27431/) (2026)
- [Before Fusion, Ask What to Keep: Contextual Calibration of Multimodal Signals](https://mlsanity.com/paper/2606.02679/) (2026)
- [Interpretable Column Annotation with LLM-Symbolized Decision Process Materialization](https://mlsanity.com/paper/2607.25228/) (2026)
- [PrivMeSA: Privacy-Aware Self-Evolving Multi-Agent System for Medicine via Local-Remote LLM Collaboration](https://mlsanity.com/paper/2609.38458/) (2026)

## Abstract (arXiv)

Knowledge graphs (KGs) provide structured evidence that can ground large language model (LLM) reasoning for knowledge-intensive question answering. However, many practical KGs are private, and sending retrieved triples or exploration traces to closed-source LLM APIs introduces leakage risk. Existing privacy treatments focus on masking entity names, but they still face four limitations: structural leakage under semantic masking, uncontrollable remote interaction, fragile multi-hop and multi-entity reasoning, and limited experience reuse for stability and efficiency. To address these issues, we propose PrivGemo, a privacy-preserving retrieval-augmented framework for KG-grounded reasoning with memory-guided exposure control. PrivGemo uses a dual-tower design to keep raw KG knowledge local while enabling remote reasoning over an anonymized view that goes beyond name masking to limit both semantic and structural exposure. PrivGemo supports multi-hop, multi-entity reasoning by retrieving anonymized long-hop paths that connect all topic entities, while keeping grounding and verification on the local KG. A hierarchical controller and a privacy-aware experience memory further reduce unnecessary exploration and remote interactions. Comprehensive experiments on six benchmarks show that PrivGemo achieves overall state-of-the-art results, outperforming the strongest baseline by up to 17.1%. Furthermore, PrivGemo enables smaller models (e.g., Qwen3-4B) to achieve reasoning performance comparable to that of GPT-4-Turbo.