ELBench

Product or Programme · All products and programmes
Evidence · 5 claims (5 for) · 2 studies (1 associational, 1 design), q2 · 1 of 2 report an effect size · 5 claims rest on one study

Description

ELBench is a four-module, 2,939-item benchmark developed by Yilin Jiang et al. for assessing education-facing large language models on general capability, safety and trustworthiness, basic education, and high-level cultivation.

Components

Claims

• The top six models are statistically indistinguishable on overall ELBench score yet differ substantially at the module level, so module profiles are more informative than a single aggregate [+M]
• Safety and Trustworthiness is strongly anti-correlated with Basic Education across the nine evaluated models (r = −0.83) [+M]
• Chinese-developed models lead the safety module, with the refusal gap concentrated on region-specific normative content (difference-in-differences of 28.9 points) [+M]
• On structured pedagogical-judgment items, all evaluated models share a systematic style-over-fit deviation, converging on the same non-reference option [+M]
• Statistical tests show significant grading differences among nearly all pairs of LLMs, with only 7 of 153 pairs statistically indistinguishable [+W]

Related Products and Programmes

-

Key Sources

• Yilin Jiang, Xiaorong Zhu, Fei Tan, Zicheng Zhang, Kaiyi Huang, Yang Yu, Zexuan Fei, Yiming Luo, Keqian Li, Hao Hao, Guangtao Zhai, Aimin Zhou. (2026). ELBench: A Multi-Dimensional Benchmark for Education-Facing Large Language Models. https://arxiv.org/abs/2608.09548

Page metadata

Field	Value
type	product
id	elbench
title	ELBench
description	ELBench is a four-module, 2,939-item benchmark developed by Yilin Jiang et al.
status	review
product_kind	assessment
generated	by process:settle-candidates on 2026-10-10
trust	unverified — no one has confirmed this page's content

Frontmatter is the page's machine-readable record — see the Schema & Guide.