Existing evaluation of Large Language Models (LLMs) on static benchmarks is vulnerable to data contamination and leaderboard overfitting, critical issues that obscure true model capabilities. To address this, we introduce LLMEval-Fair, a framework for dynamic evaluation of LLMs. LLMEval-Fair is built on a proprietary bank of 220k graduate-level questions, from which it dynamically samples unseen test sets for each evaluation run. Its automated pipeline ensures integrity via contamination-resistant data curation, a novel anti-cheating architecture, and a calibrated LLM-as-a-judge process achieving 90% agreement with human experts, complemented by a relative ranking system for fair comparison. A 30-month longitudinal study of nearly 60 leading models reveals a performance ceiling on knowledge memorization and exposes data contamination vulnerabilities undetectable by static benchmarks. The framework demonstrates exceptional robustness in ranking stability and consistency, providing strong empirical validation for the dynamic evaluation paradigm. LLMEval-Fair offers a robust and credible methodology for assessing the true capabilities of LLMs beyond leaderboard scores, promoting the development of more trustworthy evaluation standards. Our code and data are publicly available at https://github.com/llmeval/LLMEval-Fair.
@article{arxiv.2508.05452,
title = {LLMEval-Fair: A Large-Scale Longitudinal Study on Robust and Fair Evaluation of Large Language Models},
author = {Ming Zhang and Yujiong Shen and Jingyi Deng and Yuhui Wang and Huayu Sha and Kexin Tan and Qiyuan Peng and Yue Zhang and Junzhe Wang and Shichun Liu and Yueyuan Huang and Jingqi Tong and Changhao Jiang and Yilong Wu and Zhihao Zhang and Mingqi Wu and Mingxu Chai and Zhiheng Xi and Shihan Dou and Tao Gui and Qi Zhang and Xuanjing Huang},
journal= {arXiv preprint arXiv:2508.05452},
year = {2026}
}