Content alignment

E-EVAL is a comprehensive evaluation benchmark specifically designed for Chinese K-12 education, consisting of 4,351 multiple-choice questions across primary, middle, and high school levels covering nine subjects (Chinese, English, Politics, History, Ethics, Physics, Chemistry, Mathematics, Geography) to assess LLM capabilities in the Chinese K-12 education domain.

EduEval is a comprehensive hierarchical benchmark for evaluating LLMs in Chinese K-12 education, comprising 24 task types with over 11,000 questions organized across six cognitive dimensions (Memorization, Understanding, Application, Reasoning, Creativity, and Ethics) based on Bloom's Taxonomy and Webb's Depth of Knowledge. The benchmark uses authentic educational materials including real exam questions, classroom dialogues, student essays, and expert-designed prompts spanning primary through high school levels.

FoundationalASSIST introduces a 1.7-million interaction K-12 educational dataset with full question text, student responses, and Common Core alignment, specifically designed to evaluate whether LLMs can perform knowledge tracing (predicting student performance) and pedagogical grounding (understanding assessment item properties). The paper evaluates four frontier LLMs on these tasks, revealing significant gaps in their ability to predict student performance and understand item discrimination.

EduAdapt introduces a benchmark dataset of nearly 48k grade-labeled QA pairs across grades 1-12 and nine science subjects to evaluate whether LLMs can adapt their responses to different grade levels. The paper evaluates multiple open-source LLMs and finds they struggle to generate developmentally appropriate responses, especially for early-grade students.

MDK12-Bench is a large-scale benchmark built from 141K real-world K-12 exam questions across six disciplines (Math, Physics, Chemistry, Biology, Geography, Information Science) with 6,225 structured knowledge points, designed to evaluate multimodal large language models on problem-solving capabilities across difficulty levels, temporal shifts, contextual shifts, and knowledge-driven reasoning. The benchmark includes a dynamic evaluation framework and knowledge-point reference-augmented generation (KP-RAG) to assess model generalization and the role of knowledge in problem-solving.

FACET is a teacher-facing LLM-based multi-agent system that generates personalized mathematics worksheets for grade 8 students by modeling learner profiles (cognitive proficiency and intrinsic motivation), adapting content through a teacher agent, and evaluating quality through an automated evaluator agent. The system was evaluated through automated agent-based assessment and exploratory feedback from K-12 in-service teachers on authentic curriculum content.

This paper evaluates three multi-agent LLM systems that embed the Knowledge-Learning-Instruction (KLI) framework to generate secondary Math and Science learning activities, comparing them against a baseline single-agent system through teacher evaluations and LLM-as-a-judge assessments using Quality Matters K-12 standards. The study demonstrates that collaborative multi-agent systems can produce more pedagogically sound, creative, and classroom-ready activities by encoding learning science principles directly into the AI architecture.

This paper presents a framework for finetuning LLMs to generate grade-appropriate educational content across six grade levels (lower elementary through adult), evaluating the pedagogical quality and age-appropriateness of AI-generated explanations using readability metrics and human evaluation with 208 participants.

This paper presents a randomized controlled trial (n=23) evaluating GPT-4's ability to personalize science texts for middle school students by profiling their learning preferences and rewriting content accordingly. The study measures student preference for personalized versus non-personalized texts, finding significant preference for aligned content.

This paper develops and evaluates a ChatGPT-based system that automatically generates personalized reading comprehension exercises (passages and multiple-choice questions) for middle school English learners in China, with extensive human evaluation by students, teachers, and native speakers demonstrating the quality meets or exceeds human-written materials.

EduBench introduces a comprehensive benchmark dataset with 18,821 data points across 9 educational scenarios and 4,000+ educational contexts (covering K-12 and higher education subjects, different difficulty levels, and multiple question types), evaluated using 12 multi-dimensional metrics covering scenario adaptation, factual/reasoning accuracy, and pedagogical application. The benchmark assesses LLM capabilities in diverse educational tasks including assignment grading, study planning, tutoring, and psychological counseling, with both human and automated evaluation.

This paper evaluates the impact of an AI-based adaptive learning Moodle plugin on learning effectiveness and engagement with 102 Moroccan high school students studying computer science, comparing their performance and satisfaction against conventional learning methods. The study measures student academic performance, engagement levels, and participant satisfaction with the adaptive system.

This paper presents an automated multilingual pipeline for generating, solving, and evaluating K-10 math problems using three commercial LLMs (GPT-4o-mini, Gemini 2.5 Flash, Qwen-plus) across English, German, and Arabic. The study reveals consistent linguistic bias, with English solutions rated highest and Arabic lowest, highlighting equity concerns in AI-based educational tools.

This paper develops EDUMATH, a system for generating math word problems aligned with K-12 education standards using LLMs, and evaluates over 11,000 generated problems with teacher annotations for solvability, accuracy, educational appropriateness, and standards alignment. The work includes a student study comparing performance and preference between LLM-generated and human-written problems.

This paper introduces MathQuest, a mathematics dataset derived from Indian 11th and 12th grade NCERT textbooks, and evaluates three large language models (LLaMA-2, WizardMath, MAmmoTH) on mathematical problem-solving tasks through fine-tuning experiments. The work benchmarks LLM performance on secondary-level mathematics content across varying complexity levels.

This paper develops and evaluates automatic readability assessment models for science education documents in Basque, Spanish, and English for secondary education (ages 12-16), using both machine learning and deep learning approaches to help teachers select appropriately leveled texts for multilingual STEM instruction. The work creates three annotated corpora (BasqueARA, Agrega2Es, Agrega2En+) and tests feature-based and transformer-based models to classify text difficulty levels.

IndoMMLU is a multi-task language understanding benchmark for Indonesian that evaluates LLMs on 14,981 questions from primary school through university entrance exams across 64 tasks, including STEM, humanities, social science, Indonesian language proficiency, and nine local languages/cultures. The benchmark tests general reasoning, content knowledge, and multilingual capabilities of LLMs like GPT-3.5, BLOOMZ, and Falcon across Indonesian education levels.

This paper introduces the VNHSGE dataset, comprising over 19,000 multiple-choice questions and 300 essays across nine subjects from Vietnamese high school graduation examinations, designed to benchmark large language models' performance against student-level competencies. The authors evaluate ChatGPT and BingChat on this dataset, comparing their performance to Vietnamese students across subjects including mathematics, literature, sciences, and humanities.

EduPlanner is an LLM-based multi-agent system that automatically generates and iteratively optimizes instructional designs (lesson plans) for mathematics curriculum, using a Skill-Tree structure to model student knowledge levels and a five-dimensional evaluation framework (CIDDP) to assess lesson plan quality.

CMMaTH is a Chinese multimodal mathematics benchmark containing 23,000 K-12 math problems (elementary to high school) designed to evaluate foundation models' ability to solve visual-mathematical problems across different grade levels and knowledge points. The paper introduces GradeGPT, an open-source evaluation tool for assessing model performance on this dataset.

This paper introduces and evaluates prompt-based metrics for measuring the difficulty level of educational texts, aiming to improve upon traditional static metrics like Flesch-Kincaid for assessing whether LLM-generated content is appropriately adapted to different education levels. The work focuses on developing better evaluation methods for measuring how well AI systems adapt educational content difficulty, combining LLM-based prompts with traditional readability metrics.

This paper introduces STANDARDIZE, a framework for aligning LLM-generated content with expert-defined educational standards (CEFR and Common Core Standards) through retrieval-style in-context learning, focusing on generating text content at appropriate reading levels for language learners.

This paper describes Kwame for Science, an AI teaching assistant deployed in West Africa that answers students' science questions by retrieving relevant passages from curated knowledge sources and past national exam questions, achieving 87.2% top-3 accuracy over 8 months with 750 users across 32 countries. The system uses SBERT for semantic similarity matching and includes automatic topic categorization of exam questions aligned to the West African Senior Secondary Certificate Examination syllabus.

This paper evaluates the ability of large language models (ChatGPT, Copilot, Claude) to personalize board games for educational use by suggesting modifications aligned with specific classroom contexts and learning objectives, comparing AI-generated suggestions against human expert recommendations.

This study compares lesson plans created by human curriculum designers versus two GenAI models (fine-tuned LLaMA-2-13b and customized GPT-4) in K-12 mathematics education, evaluating educator preferences across instructional components (warm-up, main task, cool-down) and grade levels. The mixed-methods analysis examines both quantitative preferences and qualitative feedback to identify strengths and limitations of AI-generated lesson plans compared to human-authored ones.

This study evaluates ChatGPT-generated resources for K-12 information literacy education using Common Sense Media's Digital Citizenship framework as a template, assessing the suitability and effectiveness of AI-generated instructional materials for teaching digital citizenship concepts.

This paper presents a systematic analysis of AI-driven database systems implemented across 47 educational institutions (K-12 and higher education) over three years, evaluating their impact on personalized adaptive learning, student engagement, and academic performance. The study examines technical integration patterns, learning outcomes, and implementation challenges of AI-enhanced learning management systems.

This paper investigates how different prompt engineering strategies affect the cognitive alignment of AI-generated quiz questions with Bloom's Taxonomy levels in OneClickQuiz, a Moodle plugin for automated quiz generation. The study evaluates three prompt variants using automated classification and human review to determine which approaches best ensure pedagogically appropriate question generation at intended cognitive levels.

This paper evaluates instruction-tuned language models (ChatGPT, BLOOMZ, FlanT5, Llama) on their ability to generate and simplify text at specified readability levels using standard metrics like Flesch-Kincaid Grade Level and CEFR, tasks commonly performed by teachers when creating educational materials.

This paper develops a theoretical framework based on Constructive Alignment and Bloom's taxonomy to guide how educational assessment should adapt when students use AI tools like LLMs, examining whether current assessment methods remain valid and how AI affects evaluation of learning objectives at different cognitive levels.

This paper proposes DKT-STDRL, a deep learning model that combines CNNs and BiLSTMs to predict student learning performance by extracting spatial and temporal features from student exercise sequences in intelligent tutoring systems. The model is evaluated on educational datasets (ASSISTments, Statics2011) to predict whether students will answer questions correctly.

COGENT is a framework for generating curriculum-aligned, grade-appropriate science reading materials using LLMs, incorporating NGSS standards, readability controls, and wonder-based learning approaches. The paper evaluates generated content across curriculum alignment, comprehensibility, and readability metrics using both LLM-as-judge and human expert analysis.

This paper introduces a GenAI-powered digital assistant that uses an 'interactive mega-prompt' system to help educators create customized lesson plans tailored to student demographics, learning objectives, and teaching styles. The tool was evaluated for effectiveness through quantitative metrics (time savings) and qualitative feedback (user satisfaction) from educators across various subjects and educational levels.

This paper fine-tunes IndoBERT to automatically classify Indonesian elementary school exam questions according to Bloom's Taxonomy cognitive levels, achieving 97% accuracy. The system aims to reduce teachers' manual workload in categorizing questions by cognitive difficulty.

This paper develops and compares multiple machine learning approaches (traditional ML, RNNs, transformers, and LLMs) to automatically classify exam questions and learning outcomes according to Bloom's Taxonomy cognitive levels using a dataset of 600 labeled sentences. The study finds that augmented Support Vector Machines achieve the best performance (94% accuracy) while deep learning models suffer from overfitting on the small dataset.

This paper develops a method to control pre-trained language models for grade-specific text simplification, predicting low-level edit operations needed to simplify texts to specific reading grade levels (grades 3-8). The work focuses on automatic text rewriting to match target grade-level readability rather than evaluating learning outcomes or pedagogical effectiveness.

This paper proposes SRC, a set-to-sequence ranking model that generates personalized learning paths by recommending sequences of learning concepts tailored to individual students' needs and target learning objectives. The method captures correlations between concepts and uses knowledge tracing to evaluate learning effects, tested on educational datasets including university-level mathematics courses.

This paper proposes the OKCE model that combines machine learning feature reduction (Elastic Net, Random Forest) with deep knowledge tracing to extract knowledge component relationships and complete knowledge-concept graphs in educational curricula. The approach aims to support adaptive and personalized learning by discovering hidden relationships among curriculum concepts based on student performance data.

This paper proposes a conceptual framework for AI-driven personalized learning pathways across K-12, higher education, and corporate training, discussing student profiling, adaptive learning path generation, and continuous assessment. It references existing systems like Summit Learning (K-12) and discusses implementation challenges including data privacy, scalability, and equity.

This is a literature review paper in Portuguese that discusses the theoretical integration of AI for personalized learning in basic education (K-12), examining opportunities, challenges, ethical concerns, and the need for teacher training. The paper does not present or evaluate any specific AI system, benchmark, dataset, or empirical study of AI's impact on student learning outcomes.

This paper presents AGREE, a dataset of 42,850 computer science learning resources (research papers, lecture notes, Wikipedia content) and evaluates their readability using 14 readability indices and 12 lexical measures, training machine learning models to assess text difficulty. The work focuses on readability assessment of CS academic and technical content for self-learners, not K-12 educational contexts.

This paper uses GPT-4 to automatically generate and tag Knowledge Components (KCs) for multiple-choice questions in higher-education Chemistry and E-Learning courses, comparing LLM-generated KCs against human-created ones and developing clustering algorithms to group similar questions. The work focuses on automating the KC tagging process to support adaptive learning systems and learning analytics.

This is a retrospective review paper that examines the potential role of AI technologies (Adaptive Learning Systems, Intelligent Tutoring Systems, virtual laboratories) in teaching science education in Nigerian secondary schools, based on analysis of secondary sources rather than empirical evaluation or benchmark development.

This paper analyzes 18 AI textbooks used in Chinese primary and secondary schools, comparing them across five dimensions including teaching arrangement, content, difficulty, and alignment with curriculum standards. The study uses content analysis and quantitative curriculum difficulty models to evaluate textbook quality but does not evaluate AI systems or their impact on student learning.

This paper presents PRIMARYAI, a story-driven game-based learning environment designed to teach AI concepts (like AI planning) to upper elementary students (ages 8-11) through inquiry-based problem solving about endangered penguins. The study evaluates student engagement and learning outcomes using a Use-Modify-Create scaffolding framework with 16 students.

This paper develops topic modeling methods (LDA, OnlineLDA) to assess complexity of Russian academic texts by analyzing topical structure parameters (number of topics, coherence, distribution, weight) across school textbooks and texts for different grade levels.

This paper describes the development of LEMI, a corpus-based digital tool that automatically assesses text complexity and provides reading-level reports for Romanian schoolchildren ages 7-11 to support literacy development. The tool creates a digital reading repository with age-appropriate texts to address Romania's high functional illiteracy rates.

This paper discusses the potential integration of ChatGPT into blended learning systems to enhance sustainability, personalization, and student engagement, but does not present empirical evaluation, benchmarks, or systematic assessment of learning outcomes. The work is conceptual and review-oriented rather than an evaluation study or benchmark.

This systematic literature review identifies drivers of personalized learning in Education 4.0 and examines how AI can support personalization through learner profiling, adaptive content, and intelligent interfaces. The paper is a theoretical review of existing literature rather than an evaluation of specific AI systems or benchmarks.

This paper proposes a theoretical model for training pre-service foreign language teachers to use artificial intelligence technologies in language instruction, covering pedagogical integration of AI tools like chatbots, voice assistants, and intelligent tutoring systems. The work focuses on teacher preparation at the university level rather than direct evaluation of AI systems or measurement of K-12 student learning outcomes.