arXiv cs.AI· Tom Quilter (University of Manchester), Anastasia Ilick (Google DeepMind), Karen Poon (MathWorks), Richard Turner (University of Cambridge)·· 10 小时前AI 评分31
基于离散变分推断的教育预测模型实现数学考试成绩预测准确率超80%
Accurate Predictions in Education with Discrete Variational Inference
AI 导读
研究团队发布迄今最大的专业阅卷形式数学考试作答开放数据集，并提出基于Item Response Theory（IRT）的概率建模框架，预测学生答题正确率超过80%，创下形式化数学试卷预测准确率新基准。其协作过滤模型引入主题级技能画像后发现，仅需单一潜在能力参数即可达到最高预测精度。核心贡献是推导并实现新型离散变分推断框架，在低数据场景下取得最高预测准确率，优于所有经典IRT与矩阵分解基线。
正文 · 原文
Abstract:One of the largest drivers of social inequality is unequal access to personal tutoring, with wealthier individuals able to afford it, while the majority cannot. Affordable, effective AI tutors offer a scalable solution. We focus on adaptive learning, predicting whether a student will answer a question correctly, a key component of any effective tutoring system. Yet many platforms struggle to achieve high prediction accuracy, especially in data-sparse settings. To address this, we release the largest open dataset of professionally marked formal mathematics exam responses to date. We introduce a probabilistic modelling framework rooted in Item Response Theory (IRT) that achieves over 80 percent accuracy, setting a new benchmark for mathematics prediction accuracy of formal exam papers. Extending this, our collaborative filtering models incorporate topic-level skill profiles, but reveal a surprising and educationally significant finding, a single latent ability parameter alone is needed to achieve the maximum predictive accuracy. Our main contribution though is deriving and implementing a novel discrete variational inference framework, achieving our highest prediction accuracy in low-data settings and outperforming all classical IRT and matrix factorisation baselines.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2509.23484 [cs.AI] |
| (or arXiv:2509.23484v3 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2509.23484 arXiv-issued DOI via DataCite |
Submission history
From: Tom Quilter Dr [view email]
[v1]
Sat, 27 Sep 2025 20:13:02 UTC (392 KB)
[v2]
Tue, 30 Sep 2025 09:43:08 UTC (392 KB)
[v3]
Tue, 10 Nov 2026 11:38:25 UTC (395 KB)
来源：arXiv cs.AI · arxiv.org