✨Can MLLMs Read Students' Minds? Unpacking Multimodal Error Analysis in Handwritten Math 📝 Summary: ScratchMath introduces a benchmark for analyzing errors in student handwritten math. It reveals MLLMs significantly lag human experts in visual and logical reasoning, but proprietary models show potential for error explanation. 🔹 Publication Date: Published on Mar 26 🔹 Paper Links: • arXiv Page: https://arxiv.org/abs/2603.24961 • PDF: https://arxiv.org/pdf/2603.24961 • Project Page: https://bbsngg.github.io/ScratchMath/ • Github: https://github.com/ai-for-edu/ScratchMath ================================== For more data science resources: ✓ https://t.me/DataScienceT #AI #DataScience #MachineLearning #HuggingFace #Research