EGS: A Complementary Safety Framework for Educational LLMs
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.
Authors
- Bo Jiang (ORCID: https://orcid.org/0009-0000-1080-3278)
- Aimin Zhou (ORCID: https://orcid.org/0000-0002-4768-5946)
- Zhengnan Guo
- Zexuan Fei
- Fei Tan
- Kaiyi Huang
- Yiming Luo
Publication Details
- Journal
- arXiv (Cornell University)
- Published
- 2026-11-17
- DOI
- https://doi.org/10.48550/arxiv.2611.22203
- Primary Topic
- Adversarial Robustness in Machine Learning
- Type
- preprint