Lune

EMNLP Findings2023

Three Questions Concerning the Use of Large Language Models to Facilitate Mathematics Learning

An-Zi Yen, Wei-Ling Hsu

摘要

Due to the remarkable language understanding and generation abilities of large language models (LLMs), their use in educational applications has been explored. However, little work has been done on investigating the pedagogical ability of LLMs in helping students to learn mathematics. In this position paper, we discuss the challenges associated with employing LLMs to enhance students' mathematical problem-solving skills by providing adaptive feedback. Apart from generating the wrong reasoning processes, LLMs can misinterpret the meaning of the question, and also exhibit difficulty in understanding the given questions' rationales when attempting to correct students' answers. Three research questions are formulated.

问 Lune

几分钟读透这篇 Paper,不用再花一下午。

Lune 能读完全文,也能读引用它的顶会 Paper,讲清它的结论、证据和局限,每一点都注明出自哪篇 Paper。不用硬啃,也不靠猜。

已有 4 篇顶会 Paper 引用了它。别等审稿人问起,先弄清它们发现了什么。

可以从这些问题问起

Lune 可以查阅

全文 · 引用它的顶会 Paper · 相关的 Paper

获取答案

免费开始,无需绑卡

脉络

  • PAL: Program-aided Language Models 引用了 Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
  • Ask-Before-Detection: Identifying and Mitigating Conformity Bias in LLM-Powered Error Detector for Math Word Problem Solutions 引用了 Stepwise Verification and Remediation of Student Reasoning Errors with Large Language Model Tutors
  • Toolformer: Language Models Can Teach Themselves to Use Tools 引用了 PAL: Program-aided Language Models
  • Chain-of-Thought Prompting Elicits Reasoning in Large Language Models 引用了 Generate & Rank: A Multi-task Framework for Math Word Problems
  • Chain-of-Thought Prompting Elicits Reasoning in Large Language Models 引用了 Training language models to follow instructions with human feedback
  • AgentCoMa: A Compositional Benchmark Mixing Commonsense and Mathematical Reasoning in Real-World Scenarios 引用了 PAL: Program-aided Language Models
  • AgentCoMa: A Compositional Benchmark Mixing Commonsense and Mathematical Reasoning in Real-World Scenarios 引用了 Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
它引用的引用它的从左到右按年份排列,引用越多,位置越高。

引用它的顶会 Paper4

  1. Relief or displacement? How teachers are negotiating generative AI's role in their professional practiceAayushi Dangol, Smriti Kotiyal, Robert Wolfe, Alex J. Bowers 等CHI 2026 · 5 次引用
  2. Stepwise Verification and Remediation of Student Reasoning Errors with Large Language Model TutorsNico Daheim, Jakub Macina, Manu Kapur, Iryna Gurevych 等EMNLP 2024 · 1 次引用
  3. AgentCoMa: A Compositional Benchmark Mixing Commonsense and Mathematical Reasoning in Real-World ScenariosLisa Alazraki, Lihu Chen, Ana Brassard, Joe Stacey 等ACL 2026 · 1 次引用
  4. Ask-Before-Detection: Identifying and Mitigating Conformity Bias in LLM-Powered Error Detector for Math Word Problem SolutionsHang Li, Tianlong Xu, Kaiqi Yang, Yucheng Chu 等ACL 2025

它引用的 Paper7

  1. Training language models to follow instructions with human feedbackLong Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida 等NeurIPS 2022 · 24,707 次引用
  2. Chain-of-Thought Prompting Elicits Reasoning in Large Language ModelsJason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma 等NeurIPS 2022 · 22,562 次引用
  3. Toolformer: Language Models Can Teach Themselves to Use ToolsTimo Schick, Jane Dwivedi-Yu, Roberto Dessì, Roberta Raileanu 等NeurIPS 2023 · 5,989 次引用
  4. PAL: Program-aided Language ModelsLuyu Gao, Aman Madaan, Shuyan Zhou, Uri Alon 等ICML 2023 · 700 次引用
  5. Generate & Rank: A Multi-task Framework for Math Word ProblemsJianhao Shen, Yichun Yin, Lin Li, Lifeng Shang 等EMNLP Findings 2021 · 88 次引用

相关 Paper

  1. Exposing the Achilles' Heel: Evaluating LLMs Ability to Handle Mistakes in Mathematical ReasoningJoykirat Singh, Akshay Uttama Nambi, Vibhav VineetACL 2025 · 10 次引用
  2. Number Cookbook: Number Understanding of Language Models and How to Improve ItHaotong Yang, Yi Hu, Shijia Kang, Zhouchen Lin 等ICLR 2025
  3. GSM-Plus: A Comprehensive Benchmark for Evaluating the Robustness of LLMs as Mathematical Problem SolversQintong Li, Leyang Cui, Xueliang Zhao, Lingpeng Kong 等ACL 2024 · 9 次引用
  4. No Need for Explanations: LLMs can implicitly learn from mistakes in-contextLisa Alazraki, Maximilian Mozes, Jon Ander Campos, Yi Chern Tan 等EMNLP 2025
  5. Large Language Models Cannot Self-Correct Reasoning YetJie Huang, Xinyun Chen, Swaroop Mishra, Huaixiu Steven Zheng 等ICLR 2024 · 858 次引用
黄昏的海面,两侧是细线勾勒的悬崖