cs.CLNov 17, 2026

EGS: A Complementary Safety Framework for Educational LLMs

Authors: Zexuan Fei, Zhengnan Guo, Yiming Luo, Kaiyi Huang, Aimin Zhou, Bo Jiang, Fei Tan

Organizations: East China Normal University, Shanghai, China · Shanghai Innovation Institute, Shanghai, China

Abstract

Large Language Models (LLMs) offer strong potential for personalized education, but adolescent-facing deployment raises risks of toxic, biased, misleading, or developmentally inappropriate content. We argue that educational safety should move beyond binary refusal: student queries are often ambiguous or curiosity-driven, requiring contextual guidance rather than simple blocking. We introduce the Educational Guardrail System (EGS), a lightweight, model-agnostic runtime framework that complements intrinsic alignment through query moderation, scenario-aware pedagogical guidance, and response moderation. EGS triages inputs by risk level, redirects gray-zone queries with category-specific educational interventions, and audits outputs before delivery. Experiments across general and education-specific benchmarks show that EGS improves moderation performance, refusal quality, and jailbreak robustness across diverse LLMs. These findings suggest that defense-in-depth guardrails can strengthen safety while preserving the pedagogical utility required in educational LLMs. Our code and model weights will be publicly available to support community adoption.

Figures & tables

Appendix figures & tables7 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. EduZone: A Framework for Evaluating LLM Safety for K-12 Students and Teachers

    Aug 3, 2026Junyeong Park, Jieun Han, Haneul Yoo +3LLM Safety BenchmarksGenerative AI in Education

  2. SHAPE: Unifying Safety, Helpfulness and Pedagogy for Educational LLMs

    Apr 24, 2026Sihang Zhao, Kangrui Yu, Youliang Yuan +2Intelligent Tutoring SystemsGenerative AI in Education

  3. ELBench: A Multi-Dimensional Benchmark for Education-Facing Large Language Models

    Aug 10, 2026Yilin Jiang, Xiaorong Zhu, Fei Tan +9LLM Safety BenchmarksLLM Evaluation