# Towards Robust Evaluation of STEM Education: Leveraging MLLMs in Project-Based Learning arXiv: https://arxiv.org/abs/2505.17050 ยท HTML: https://mlsanity.com/paper/2505.17050/ ## Key facts - Cited by 7 papers in the map (data through 14 Oct 2026). - Builds on 28 of its 46 references that are in the map. - First posted on arXiv 16 May 2025. ## Builds on (most cited first, 28 of 28) - [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021) - [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021) - [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020) - [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025) - [Measuring Mathematical Problem Solving With the MATH Dataset](https://mlsanity.com/paper/2103.03874/) (2021) - [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024) - [Improved Baselines with Visual Instruction Tuning](https://mlsanity.com/paper/2310.03744/) (2023) - [Program Synthesis with Large Language Models](https://mlsanity.com/paper/2108.07732/) (2021) - [MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGI](https://mlsanity.com/paper/2311.16502/) (2023) - [Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering](https://mlsanity.com/paper/2209.09513/) (2022) - [WorldSimBench: Towards Video Generation Models as World Simulators](https://mlsanity.com/paper/2410.18072/) (2024) - [Phi-4 Technical Report](https://mlsanity.com/paper/2412.08905/) (2024) - [MDAgents: An Adaptive Collaboration of LLMs for Medical Decision-Making](https://mlsanity.com/paper/2404.15155/) (2024) - [DynaMath: A Dynamic Visual Benchmark for Evaluating Mathematical Reasoning Robustness of Vision Language Models](https://mlsanity.com/paper/2411.00836/) (2024) - [FinCon: A Synthesized LLM Multi-Agent System with Conceptual Verbal Reinforcement for Enhanced Financial Decision Making](https://mlsanity.com/paper/2407.06567/) (2024) - [MP5: A Multi-modal Open-ended Embodied System in Minecraft via Active Perception](https://mlsanity.com/paper/2312.07472/) (2023) - [Universal Vulnerabilities in Large Language Models: Backdoor Attacks for In-context Learning](https://mlsanity.com/paper/2401.05949/) (2024) - [RestoreAgent: Autonomous Image Restoration Agent via Multimodal Large Language Models](https://mlsanity.com/paper/2407.18035/) (2024) - [EduAgent: Generative Student Agents in Learning](https://mlsanity.com/paper/2404.07963/) (2024) - [NovelQA: Benchmarking Question Answering on Documents Exceeding 200K Tokens](https://mlsanity.com/paper/2403.12766/) (2024) - [RoboFactory: Exploring Embodied Agent Collaboration with Compositional Constraints](https://mlsanity.com/paper/2503.16408/) (2025) - [Unlearning Backdoor Attacks for LLMs with Weak-to-Strong Knowledge Distillation](https://mlsanity.com/paper/2410.14425/) (2024) - [FreestyleRet: Retrieving Images from Style-Diversified Queries](https://mlsanity.com/paper/2312.02428/) (2023) - [Measuring Vision-Language STEM Skills of Neural Models](https://mlsanity.com/paper/2402.17205/) (2024) - [KDMCSE: Knowledge Distillation Multimodal Sentence Embeddings with Adaptive Angular margin Contrastive Learning](https://mlsanity.com/paper/2403.17486/) (2024) - [Uni-Retrieval: A Multi-Style Retrieval Framework for STEM's Education](https://mlsanity.com/paper/2502.05863/) (2025) - [Exploring Cognitive and Aesthetic Causality for Multimodal Aspect-Based Sentiment Analysis](https://mlsanity.com/paper/2504.15848/) (2025) - [Enhancing Multimodal Entity Linking with Jaccard Distance-based Conditional Contrastive Learning and Contextual Visual Augmentation](https://mlsanity.com/paper/2501.14166/) (2025) ## Built on by (most cited first, 7 of 7) - [VIKI-R: Coordinating Embodied Multi-Agent Cooperation via Reinforcement Learning](https://mlsanity.com/paper/2506.09049/) (2025) - [Beyond Vulnerabilities: A Survey of Adversarial Attacks as Both Threats and Defenses in Computer Vision Systems](https://mlsanity.com/paper/2508.01845/) (2025) - [Affective-ROPTester: Capability and Bias Analysis of LLMs in Predicting Retinopathy of Prematurity](https://mlsanity.com/paper/2507.05816/) (2025) - [From Query to Explanation: Uni-RAG for Multi-Modal Retrieval-Augmented Learning in STEM](https://mlsanity.com/paper/2507.03868/) (2025) - [Investigating Vulnerabilities and Defenses Against Audio-Visual Attacks: A Comprehensive Survey Emphasizing Multimodal Models](https://mlsanity.com/paper/2506.11521/) (2025) - [A Dialogue-Based Framework for Correcting Multimodal Errors in AI-Assisted STEM Education](https://mlsanity.com/paper/2605.04131/) (2026) - [Human-AI Co-Mentorship in Project-Based Learning: A Case Study in Financial Forecasting](https://mlsanity.com/paper/2605.05144/) (2026) ## Abstract (arXiv) Project-Based Learning (PBL) involves a variety of highly correlated multimodal data, making it a vital educational approach within STEM disciplines. With the rapid development of multimodal large language models (MLLMs), researchers have begun exploring their potential to enhance tasks such as information retrieval, knowledge comprehension, and data generation in educational settings. However, existing benchmarks fall short in providing both a free-form output structure and a rigorous human expert validation process, limiting their effectiveness in evaluating real-world educational tasks. Additionally, few methods have developed automated pipelines to assist with the complex responsibilities of teachers leveraging MLLMs, largely due to model hallucination and instability, which lead to unreliable implementation. To address this gap, we introduce PBLBench, a novel benchmark designed to evaluate complex reasoning grounded in domain-specific knowledge and long-context understanding, thereby challenging models with tasks that closely resemble those handled by human experts. To establish reliable ground truth, we adopt the Analytic Hierarchy Process (AHP), utilizing expert-driven pairwise comparisons to derive structured and weighted evaluation criteria. We assess the performance of 15 leading MLLMs/LLMs using PBLBench and demonstrate that even the most advanced models achieve only 59% rank accuracy, underscoring the significant challenges presented by this benchmark. We believe PBLBench will serve as a catalyst for the development of more capable AI agents, ultimately aiming to alleviate teacher workload and enhance educational productivity.