Edu-Values: Towards Evaluating the Chinese Education Values of Large Language Models
Not on the map. Why?
Abstract (arXiv)
In this paper, we present Edu-Values, the first Chinese education values evaluation benchmark that includes seven core values: professional philosophy, teachers' professional ethics, education laws and regulations, cultural literacy, educational knowledge and skills, basic competencies and subject knowledge. We meticulously design 1,418 questions, covering multiple-choice, multi-modal question answering, subjective analysis, adversarial prompts, and Chinese traditional culture (short answer) questions. We conduct human feedback based automatic evaluation over 21 state-of-the-art (SoTA) LLMs, and highlight three main findings: (1) due to differences in educational culture, Chinese LLMs outperform English LLMs, with Qwen 2 ranking the first with a score of 81.37; (2) LLMs often struggle with teachers' professional ethics and professional philosophy; (3) leveraging Edu-Values to build an external knowledge repository for RAG significantly improves LLMs' alignment. This demonstrates the effectiveness of the proposed benchmark.
Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Builds on 31 of its 45 references that are in the map. First posted on arXiv 19 Sep 2024.
Builds on
31 of 45 references.
References · 31
- 2022Training language models to follow instructions with human feedbackcited by 15,712
- 2024The Llama 3 Herd of Modelscited by 14,192
- 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 12,865
- 2020Measuring Massive Multitask Language Understandingcited by 7,768
- 2023Qwen Technical Reportcited by 3,273
- 2024Gemini 1.5: Unlocking multimodal understanding across millions of tokens of contextcited by 3,175
- 2022Constitutional AI: Harmlessness from AI Feedbackcited by 2,950
- 2024Qwen2 Technical Reportcited by 2,076
- 2022Large Language Models Encode Clinical Knowledgecited by 1,807
- 2020RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Modelscited by 1,235
- 2024ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Toolscited by 1,096
- 2024DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Modelcited by 1,093
19 more
- 2021GLM: General Language Model Pretraining with Autoregressive Blank Infillingcited by 1,026
- 2021A General Language Assistant as a Laboratory for Alignmentcited by 1,023
- 2020Aligning AI With Shared Human Valuescited by 706
- 2024Yi: Open Foundation Models by 01.AIcited by 706
- 2021BBQ: A Hand-Built Bias Benchmark for Question Answeringcited by 683
- 2023Multi-step Jailbreaking Privacy Attacks on ChatGPTcited by 320
- 2021ERNIE 3.0: Large-scale Knowledge Enhanced Pre-training for Language Understanding and Generationcited by 258
- 2023Evaluating the Performance of Large Language Models on GAOKAO Benchmarkcited by 182
- 2023Chatlaw: A Multi-Agent Legal Assistant based on a Role-Aligned Mixture-of-Experts Architecturecited by 119
- 2017Deceiving Google's Perspective API Built for Detecting Toxic Commentscited by 99
- 2023XuanYuan 2.0: A Large Chinese Financial Chat Model with Hundreds of Billions Parameterscited by 92
- 2023CValues: Measuring the Values of Chinese Large Language Models from Safety to Responsibilitycited by 80
- 2023EduChat: A Large-Scale Language Model-based Chatbot System for Intelligent Educationcited by 80
- 2022COLD: A Benchmark for Chinese Offensive Language Detectioncited by 74
- 2023FinGPT: Instruction Tuning Benchmark for Open-Source Large Language Models in Financial Datasetscited by 68
- 2023Flames: Benchmarking Value Alignment of LLMs in Chinesecited by 35
- 2023M3KE: A Massive Multi-Level Multi-Subject Knowledge Evaluation Benchmark for Chinese Large Language Modelscited by 30
- 2023Think Twice: A Human-like Two-stage Conversational Agent for Emotional Response Generationcited by 14
- 2022Aligning Recommendation and Conversation via Dual Imitationcited by 6
Built on by
Citing papers per year: 2025: 4; 2026: 1.
5 papers.
- 2025Evaluate Bias without Manual Test Sets: A Concept Representation Perspective for LLMscited by 5
- 2025Unified Defense for Large Language Models against Jailbreak and Fine-Tuning Attacks in Educationcited by 5
- 2025NurValues: Real-World Nursing Values Evaluation for Large Language Models in Clinical Contextcited by 0
- 2025PEMUTA: Pedagogically-Enriched Multi-Granular Undergraduate Thesis Assessmentcited by 0
- 2026ADE: Agentic Data Evolution Framework for Human-Centered Objectivescited by 0
- 2026ADE: Agentic Data Evolution Framework for Human-Centered Objectives24 Aug 2026
- 2025Unified Defense for Large Language Models against Jailbreak and Fine-Tuning Attacks in Education18 Nov 2025
- 2025PEMUTA: Pedagogically-Enriched Multi-Granular Undergraduate Thesis Assessment25 Jul 2025
- 2025Evaluate Bias without Manual Test Sets: A Concept Representation Perspective for LLMs21 May 2025
- 2025NurValues: Real-World Nursing Values Evaluation for Large Language Models in Clinical Context13 May 2025
Similar, not cited
Close in wording, no citation either way
- 2024E-EVAL: A Comprehensive Chinese K-12 Education Evaluation Benchmark for Large Language Models
- 2025OmniEduBench: A Comprehensive Chinese Benchmark for Evaluating Large Language Models in Education
- 2023C-Eval: A Multi-Level Multi-Discipline Chinese Evaluation Suite for Foundation Models
- 2025Benchmarking Multi-National Value Alignment for Large Language Models
- 2025Analysis of instruction-based LLMs' capabilities to score and judge text-input problems in an academic setting
- 2024CMMU: A Benchmark for Chinese Multi-modal Multi-type Question Understanding and Reasoning
- 2025Cultural Value Alignment in Large Language Models: A Prompt-based Analysis of Schwartz Values in Gemini, ChatGPT, and DeepSeek
- 2026On the Credibility of Evaluating LLMs using Survey Questions