Saved in:
Bibliographic Details
Main Authors: Zhou, Yuhang, Chen, Xutian, Cao, Yixin, Ni, Yuchen, He, Yu, Tian, Siyu, Liu, Xiang, Zhang, Jian, Ji, Chuanjun, Ye, Guangnan, Qiu, Xipeng
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2505.12259
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910951298564096
author Zhou, Yuhang
Chen, Xutian
Cao, Yixin
Ni, Yuchen
He, Yu
Tian, Siyu
Liu, Xiang
Zhang, Jian
Ji, Chuanjun
Ye, Guangnan
Qiu, Xipeng
author_facet Zhou, Yuhang
Chen, Xutian
Cao, Yixin
Ni, Yuchen
He, Yu
Tian, Siyu
Liu, Xiang
Zhang, Jian
Ji, Chuanjun
Ye, Guangnan
Qiu, Xipeng
contents Recent progress in large language models (LLMs) has outpaced the development of effective evaluation methods. Traditional benchmarks rely on task-specific metrics and static datasets, which often suffer from fairness issues, limited scalability, and contamination risks. In this paper, we introduce Teach2Eval, an indirect evaluation framework inspired by the Feynman Technique. Instead of directly testing LLMs on predefined tasks, our method evaluates a model's multiple abilities to teach weaker student models to perform tasks effectively. By converting open-ended tasks into standardized multiple-choice questions (MCQs) through teacher-generated feedback, Teach2Eval enables scalable, automated, and multi-dimensional assessment. Our approach not only avoids data leakage and memorization but also captures a broad range of cognitive abilities that are orthogonal to current benchmarks. Experimental results across 26 leading LLMs show strong alignment with existing human and model-based dynamic rankings, while offering additional interpretability for training guidance.
format Preprint
id arxiv_https___arxiv_org_abs_2505_12259
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Teach2Eval: An Indirect Evaluation Method for LLM by Judging How It Teaches
Zhou, Yuhang
Chen, Xutian
Cao, Yixin
Ni, Yuchen
He, Yu
Tian, Siyu
Liu, Xiang
Zhang, Jian
Ji, Chuanjun
Ye, Guangnan
Qiu, Xipeng
Computation and Language
Recent progress in large language models (LLMs) has outpaced the development of effective evaluation methods. Traditional benchmarks rely on task-specific metrics and static datasets, which often suffer from fairness issues, limited scalability, and contamination risks. In this paper, we introduce Teach2Eval, an indirect evaluation framework inspired by the Feynman Technique. Instead of directly testing LLMs on predefined tasks, our method evaluates a model's multiple abilities to teach weaker student models to perform tasks effectively. By converting open-ended tasks into standardized multiple-choice questions (MCQs) through teacher-generated feedback, Teach2Eval enables scalable, automated, and multi-dimensional assessment. Our approach not only avoids data leakage and memorization but also captures a broad range of cognitive abilities that are orthogonal to current benchmarks. Experimental results across 26 leading LLMs show strong alignment with existing human and model-based dynamic rankings, while offering additional interpretability for training guidance.
title Teach2Eval: An Indirect Evaluation Method for LLM by Judging How It Teaches
topic Computation and Language
url https://arxiv.org/abs/2505.12259