ZPD-SCA: Unveiling the Blind Spots of LLMs in Assessing Students' Cognitive Abilities

TL;DR

This paper introduces ZPD-SCA, a benchmark for evaluating large language models' ability to assess Chinese reading difficulty aligned with students' developmental stages, revealing current limitations and biases in LLMs' educational assessments.

Contribution

The paper presents ZPD-SCA, the first comprehensive benchmark for Chinese reading comprehension difficulty, and evaluates LLMs' performance, highlighting their emerging abilities and existing biases in educational assessment.

Findings

LLMs perform poorly in zero-shot scenarios for reading difficulty assessment.

In-context examples significantly improve LLM performance.

Models exhibit biases and genre-based performance variations.

Abstract

Large language models (LLMs) have demonstrated potential in educational applications, yet their capacity to accurately assess the cognitive alignment of reading materials with students' developmental stages remains insufficiently explored. This gap is particularly critical given the foundational educational principle of the Zone of Proximal Development (ZPD), which emphasizes the need to match learning resources with Students' Cognitive Abilities (SCA). Despite the importance of this alignment, there is a notable absence of comprehensive studies investigating LLMs' ability to evaluate reading comprehension difficulty across different student age groups, especially in the context of Chinese language education. To fill this gap, we introduce ZPD-SCA, a novel benchmark specifically designed to assess stage-level Chinese reading comprehension difficulty. The benchmark is annotated by 60 Special Grade teachers, a group that represents the top 0.15% of all in-service teachers nationwide. Experimental results reveal that LLMs perform poorly in zero-shot learning scenarios, with Qwen-max and GLM even falling below the probability of random guessing. When provided with in-context examples, LLMs performance improves substantially, with some models achieving nearly double the accuracy of their zero-shot baselines. These results reveal that LLMs possess emerging abilities to assess reading difficulty, while also exposing limitations in their current training for educationally aligned judgment. Notably, even the best-performing models display systematic directional biases, suggesting difficulties in accurately aligning material difficulty with SCA. Furthermore, significant variations in model performance across different genres underscore the complexity of task. We envision that ZPD-SCA can provide a foundation for evaluating and improving LLMs in cognitively aligned educational applications.