# LLMs cannot spot math errors, even when allowed to peek into the solution

arXiv: https://arxiv.org/abs/2509.01395 · HTML: https://mlsanity.com/paper/2509.01395/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 26 references that are in the map.
- First posted on arXiv 1 Sep 2025.

## Builds on (most cited first, 13 of 13)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [BERTScore: Evaluating Text Generation with BERT](https://mlsanity.com/paper/1904.09675/) (2019)
- [Measuring Mathematical Problem Solving With the MATH Dataset](https://mlsanity.com/paper/2103.03874/) (2021)
- [Let's Verify Step by Step](https://mlsanity.com/paper/2305.20050/) (2023)
- [Qwen2.5-Math Technical Report: Toward Mathematical Expert Model via Self-Improvement](https://mlsanity.com/paper/2409.12122/) (2024)
- [MathDial: A Dialogue Tutoring Dataset with Rich Pedagogical Properties Grounded in Math Reasoning Problems](https://mlsanity.com/paper/2305.14536/) (2023)
- [MR-GSM8K: A Meta-Reasoning Benchmark for Large Language Model Evaluation](https://mlsanity.com/paper/2312.17080/) (2023)
- [Stepwise Verification and Remediation of Student Reasoning Errors with Large Language Model Tutors](https://mlsanity.com/paper/2407.09136/) (2024)
- [What Makes Math Word Problems Challenging for LLMs?](https://mlsanity.com/paper/2403.11369/) (2024)
- [MIRAGE-Bench: Automatic Multilingual Benchmark Arena for Retrieval-Augmented Generation Systems](https://mlsanity.com/paper/2410.13716/) (2024)
- [Ask-Before-Detection: Identifying and Mitigating Conformity Bias in LLM-Powered Error Detector for Math Word Problem Solutions](https://mlsanity.com/paper/2412.16838/) (2024)

## Built on by (most cited first, 7 of 7)

- [REFLECT: Intervention-Supported Error Attribution for Silent Failures in LLM Agent Traces](https://mlsanity.com/paper/2606.09071/) (2026)
- [When Verification Hurts: Asymmetric Effects of Multi-Agent Feedback in Logic Proof Tutoring](https://mlsanity.com/paper/2603.27076/) (2026)
- [The Aftermath of DrawEduMath: Vision Language Models Underperform with Struggling Students and Misdiagnose Errors](https://mlsanity.com/paper/2603.00925/) (2026)
- [HyperGuide: Hyperbolic Guidance for Efficient Multi-Step Reasoning in Large Language Models](https://mlsanity.com/paper/2605.24140/) (2026)
- [Incentives Of EdTech: A Systematic Review Of EduNLP Research](https://mlsanity.com/paper/2606.13691/) (2026)
- [When Tokenization is Secretly Output Supervision](https://mlsanity.com/paper/2609.01386/) (2026)
- [Correct Diagnosis, Better Feedback: A Symbolic-Verifier for Faithful LLM Tutoring Feedback in Logic Proofs](https://mlsanity.com/paper/2609.22553/) (2026)

## Abstract (arXiv)

Large language models (LLMs) demonstrate remarkable performance on math word problems, yet they have been shown to struggle with meta-reasoning tasks such as identifying errors in student solutions. In this work, we investigate the challenge of locating the first error step in stepwise solutions using two error reasoning datasets: VtG and PRM800K. Our experiments show that state-of-the-art LLMs struggle to locate the first error step in student solutions even when given access to the reference solution. To that end, we propose an approach that generates an intermediate corrected student solution, aligning more closely with the original student's solution, which helps improve performance.