# SHAPE: Unifying Safety, Helpfulness and Pedagogy for Educational LLMs

arXiv: https://arxiv.org/abs/2604.22134 · HTML: https://mlsanity.com/paper/2604.22134/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in Safety alignment, refusal and guardrails, one of 12 directions in LLM safety and security.
- Builds on 20 of its 36 references that are in the map.
- First posted on arXiv 24 Apr 2026.

## Placed in

- LLM safety and security / [Safety alignment, refusal and guardrails](https://mlsanity.com/safety/safety-alignment-refusal-and-guardrails/)

## Builds on (most cited first, 20 of 20)

- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Jailbroken: How Does LLM Safety Training Fail?](https://mlsanity.com/paper/2307.02483/) (2023)
- [GPT-4 Is Too Smart To Be Safe: Stealthy Chat with LLMs via Cipher](https://mlsanity.com/paper/2308.06463/) (2023)
- [Deep Knowledge Tracing](https://mlsanity.com/paper/1506.05908/) (2015)
- [Your Brain on ChatGPT: Accumulation of Cognitive Debt when Using an AI Assistant for Essay Writing Task](https://mlsanity.com/paper/2506.08872/) (2025)
- [Training Production Language Models without Memorizing User Data](https://mlsanity.com/paper/2009.10031/) (2020)
- [Enhancing LLM-Based Feedback: Insights from Intelligent Tutoring Systems and the Learning Sciences](https://mlsanity.com/paper/2405.04645/) (2024)
- [Enhancing Personalized Multi-Turn Dialogue with Curiosity Reward](https://mlsanity.com/paper/2504.03206/) (2025)
- [Towards Modeling Learner Performance with Large Language Models](https://mlsanity.com/paper/2403.14661/) (2024)
- [Unmasking Digital Falsehoods: A Comparative Analysis of LLM-Based Misinformation Detection Strategies](https://mlsanity.com/paper/2503.00724/) (2025)
- [From Superficial Outputs to Superficial Learning: Risks of Large Language Models in Education](https://mlsanity.com/paper/2509.21972/) (2025)
- [Knowledge Graphs are Implicit Reward Models: Path-Derived Signals Enable Compositional Reasoning](https://mlsanity.com/paper/2601.15160/) (2026)
- [PAIGE: Examining Learning Outcomes and Experiences with Personalized AI-Generated Educational Podcasts](https://mlsanity.com/paper/2409.04645/) (2024)
- [AI Meets the Classroom: When Do Large Language Models Harm Learning?](https://mlsanity.com/paper/2409.09047/) (2024)
- [CoDAE: Adapting Large Language Models for Education via Chain-of-Thought Data Augmentation](https://mlsanity.com/paper/2508.08386/) (2025)
- [Unified Defense for Large Language Models against Jailbreak and Fine-Tuning Attacks in Education](https://mlsanity.com/paper/2511.14423/) (2025)
- [Rooted Absorbed Prefix Trajectory Balance with Submodular Replay for GFlowNet Training](https://mlsanity.com/paper/2603.00454/) (2026)
- [Securing Educational LLMs: A Generalised Taxonomy of Attacks on LLMs and DREAD Risk Assessment](https://mlsanity.com/paper/2508.08629/) (2025)
- [Explore-on-Graph: Incentivizing Autonomous Exploration of Large Language Models on Knowledge Graphs with Path-refined Reward Modeling](https://mlsanity.com/paper/2602.21728/) (2026)
- [Beyond Static Question Banks: Dynamic Knowledge Expansion via LLM-Automated Graph Construction and Adaptive Generation](https://mlsanity.com/paper/2602.00020/) (2026)

## Built on by (most cited first, 3 of 3)

- [Mitigating Scaffolding Collapse in Socratic Tutors via Representation Alignment](https://mlsanity.com/paper/2607.19371/) (2026)
- [Auditable Release Control for Pedagogical Leakage in LLM Tutors](https://mlsanity.com/paper/2608.00515/) (2026)
- [ELBench: A Multi-Dimensional Benchmark for Education-Facing Large Language Models](https://mlsanity.com/paper/2608.09548/) (2026)

## Abstract (arXiv)

Large Language Models (LLMs) have been widely explored in educational scenarios. We identify a critical vulnerability in current educational LLMs, pedagogical jailbreaks, where students use answer-inducing prompts to elicit solutions rather than scaffolded instructions. To enable systematic study, we unify and formalize safe, helpful, and pedagogical behaviors with a knowledge-mastery graph and introduce SHAPE, a benchmark of 9,087 student-question pairs for evaluating tutoring behavior under adversarial pressure. We propose a graph-augmented tutoring pipeline that infers prerequisite concepts from queries, identifies mastery gaps, and routes generation between instructing and problem-solving via explicit gating. Experiments across multiple LLMs show that our method yields significantly improved safety under two pedagogical jailbreak settings, while maintaining near-ceiling helpfulness under the same evaluation protocol. Our code and data are available at https://github.com/MAPS-research/SHaPE