EMNLP: Educator-role Moral and Normative Large Language Models Profiling
This work addresses the problem of ensuring ethical and psychological alignment in educational AI for developers and educators, though it is incremental as it extends existing scales and benchmarks.
The paper tackles the lack of psychological and ethical evaluation in Large Language Models (LLMs) simulating professional roles by introducing EMNLP, a framework for profiling educator-role LLMs, and finds that teacher-role LLMs show more idealized personalities than humans, excel in abstract moral reasoning but struggle with emotional complexity, with stronger reasoning models being more vulnerable to harmful prompts.
Simulating Professions (SP) enables Large Language Models (LLMs) to emulate professional roles. However, comprehensive psychological and ethical evaluation in these contexts remains lacking. This paper introduces EMNLP, an Educator-role Moral and Normative LLMs Profiling framework for personality profiling, moral development stage measurement, and ethical risk under soft prompt injection. EMNLP extends existing scales and constructs 88 teacher-specific moral dilemmas, enabling profession-oriented comparison with human teachers. A targeted soft prompt injection set evaluates compliance and vulnerability in teacher SP. Experiments on 14 LLMs show teacher-role LLMs exhibit more idealized and polarized personalities than human teachers, excel in abstract moral reasoning, but struggle with emotionally complex situations. Models with stronger reasoning are more vulnerable to harmful prompt injection, revealing a paradox between capability and safety. The model temperature and other hyperparameters have limited influence except in some risk behaviors. This paper presents the first benchmark to assess ethical and psychological alignment of teacher-role LLMs for educational AI. Resources are available at https://e-m-n-l-p.github.io/.