Summary of Why Llms Are Bad at Synthetic Table Generation (and What to Do About It), by Shengzhe Xu et al.
Why LLMs Are Bad at Synthetic Table Generation (and what to do about it)
by Shengzhe Xu, Cho-Ting Lee, Mandar Sharma, Raquib Bin Yousuf, Nikhil Muralidhar, Naren Ramakrishnan
First submitted to arxiv on: 20 Jun 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
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 In this research paper, the authors investigate synthetic table generation using Large Language Models (LLMs). They highlight the importance of synthetic data in Machine Learning (ML) pipelines, particularly in business and science applications. However, they show that current LLMs are inadequate for generating synthetic tables due to their autoregressive nature and the random order permutation used during fine-tuning. The authors propose a solution by making LLMs permutation-aware, which can mitigate these limitations. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Synthetic data generation is crucial in ML pipelines to augment training data, replace sensitive information, or power advanced platforms. While LLMs are improving for synthetic data generation, table synthesis remains under-explored. The authors show that current LLMs fail to generate realistic tables because of their autoregressive nature and random order permutation during fine-tuning. They suggest making LLMs permutation-aware to overcome these limitations. |
Keywords
» Artificial intelligence » Autoregressive » Fine tuning » Machine learning » Synthetic data
