Guardado en:
Detalles Bibliográficos
Autores principales: Guo, Shuang, Li, Zihui
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:https://arxiv.org/abs/2512.03694
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866917121981677568
author Guo, Shuang
Li, Zihui
author_facet Guo, Shuang
Li, Zihui
contents Multi-Agent Systems (MAS) with large language models (LLMs) enable personalized education but risk leaking minors personally identifiable information (PII) via unstructured dialogue. Existing privacy methods struggle to balance security and utility: role-based access control fails on unstructured text, while naive masking destroys pedagogical context. We propose SRPG, a privacy guard for educational MAS, using a Dual-Stream Reconstruction Mechanism: a strict sanitization stream ensures zero PII leakage, and a context reconstruction stream (LLM driven) recovers mathematical logic. This decouples instructional content from private data, preserving teaching efficacy. Tests on MathDial show SRPG works across models; with GPT-4o, it achieves 0.0000 Attack Success Rate (ASR) (zero leakage) and 0.8267 Exact Match, far outperforming the zero trust Pure LLM baseline (0.2138). SRPG effectively protects minors privacy without sacrificing mathematical instructional quality.
format Preprint
id arxiv_https___arxiv_org_abs_2512_03694
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SRPG: Semantically Reconstructed Privacy Guard for Zero-Trust Privacy in Educational Multi-Agent Systems
Guo, Shuang
Li, Zihui
Multiagent Systems
Multi-Agent Systems (MAS) with large language models (LLMs) enable personalized education but risk leaking minors personally identifiable information (PII) via unstructured dialogue. Existing privacy methods struggle to balance security and utility: role-based access control fails on unstructured text, while naive masking destroys pedagogical context. We propose SRPG, a privacy guard for educational MAS, using a Dual-Stream Reconstruction Mechanism: a strict sanitization stream ensures zero PII leakage, and a context reconstruction stream (LLM driven) recovers mathematical logic. This decouples instructional content from private data, preserving teaching efficacy. Tests on MathDial show SRPG works across models; with GPT-4o, it achieves 0.0000 Attack Success Rate (ASR) (zero leakage) and 0.8267 Exact Match, far outperforming the zero trust Pure LLM baseline (0.2138). SRPG effectively protects minors privacy without sacrificing mathematical instructional quality.
title SRPG: Semantically Reconstructed Privacy Guard for Zero-Trust Privacy in Educational Multi-Agent Systems
topic Multiagent Systems
url https://arxiv.org/abs/2512.03694