Date
Publisher
arXiv
Cognitive Diagnosis (CD) has become a critical task in AI-empowered
education, supporting personalized learning by accurately assessing students'
cognitive states. However, traditional CD models often struggle in cold-start
scenarios due to the lack of student-exercise interaction data. Recent
NLP-based approaches leveraging pre-trained language models (PLMs) have shown
promise by utilizing textual features but fail to fully bridge the gap between
semantic understanding and cognitive profiling. In this work, we propose
Language Models as Zeroshot Cognitive Diagnosis Learners (LMCD), a novel
framework designed to handle cold-start challenges by harnessing large language
models (LLMs). LMCD operates via two primary phases: (1) Knowledge Diffusion,
where LLMs generate enriched contents of exercises and knowledge concepts
(KCs), establishing stronger semantic links; and (2) Semantic-Cognitive Fusion,
where LLMs employ causal attention mechanisms to integrate textual information
and student cognitive states, creating comprehensive profiles for both students
and exercises. These representations are efficiently trained with off-the-shelf
CD models. Experiments on two real-world datasets demonstrate that LMCD
significantly outperforms state-of-the-art methods in both exercise-cold and
domain-cold settings. The code is publicly available at
https://github.com/TAL-auroraX/LMCD
Study at a glance
- Question: Can large language models assess students' learning when little or no prior student practice data exists?
- Setting and data: Two real-world datasets.
- Method: Computational experiment testing a new AI framework against existing methods on two datasets.
Why it matters
Assessing students' cognitive states supports personalized learning, but traditional cognitive diagnosis models struggle in cold-start scenarios lacking interaction data. Furthermore, recent language model approaches fail to fully bridge the gap between semantic understanding and cognitive profiling.
Key findings
- Across two real-world datasets, a new language-model framework diagnosed student knowledge significantly more accurately than existing tools when past student performance data was unavailable.
- The tool maintained superior diagnostic accuracy both when evaluating completely new learning exercises and across unfamiliar subject domains.
- Earlier language models struggled to assess student learning accurately because they could not effectively connect exercise text with student cognitive profiles.
Summary drafted with AI from the paper’s abstract and may contain errors. See the full paper for details.
What is the application?
Why use AI?
Study design
