Summary of Mathgenie: Generating Synthetic Data with Question Back-translation For Enhancing Mathematical Reasoning Of Llms, by Zimu Lu et al.
MathGenie: Generating Synthetic Data with Question Back-translation for Enhancing Mathematical Reasoning of LLMs
by Zimu Lu, Aojun Zhou, Houxing Ren, Ke Wang, Weikang Shi, Junting Pan, Mingjie Zhan, Hongsheng Li
First submitted to arxiv on: 26 Feb 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary MathGenie is a novel method for generating diverse and reliable math problems from a small-scale problem-solution dataset called seed data. The approach involves augmenting ground-truth solutions, training a back-translation model to translate the augmented solutions into new questions, and then generating code-integrated solutions for these questions. To verify the correctness of these solutions, the researchers employed a rationale-based strategy. They trained various pre-trained models, ranging from 7B to 70B, on the newly curated data to test the effectiveness of their proposed augmentation technique, resulting in a family of models known as MathGenieLM. These models consistently outperformed previous open-source models across five representative mathematical reasoning datasets, achieving state-of-the-art performance. Specifically, MathGenieLM-InternLM2 achieved an accuracy of 87.7% on GSM8K and 55.7% on MATH. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary MathGenie is a new way to create math problems that are fun and challenging! It uses a special process to turn answers into new questions, and then makes sure the solutions are correct by using a clever strategy. The team trained many language models on this data to see how well they did, and the results were amazing! They even beat some really smart computers in a math competition! |
Keywords
» Artificial intelligence » Translation
