# DataVinci: Learning Syntactic and Semantic String Repairs

arXiv: https://arxiv.org/abs/2308.10922 · HTML: https://mlsanity.com/paper/2308.10922/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 39 references that are in the map.
- First posted on arXiv 21 Aug 2023.

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

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [CodeT5: Identifier-aware Unified Pre-trained Encoder-Decoder Models for Code Understanding and Generation](https://mlsanity.com/paper/2109.00859/) (2021)
- [Can Foundation Models Wrangle Your Data?](https://mlsanity.com/paper/2205.09911/) (2022)
- [HoloClean: Holistic Data Repairs with Probabilistic Inference](https://mlsanity.com/paper/1702.00820/) (2017)
- [Sherlock: A Deep Learning Approach to Semantic Data Type Detection](https://mlsanity.com/paper/1905.10688/) (2019)
- [HoloDetect: Few-Shot Learning for Error Detection](https://mlsanity.com/paper/1904.02285/) (2019)
- [Multi-modal Program Inference: a Marriage of Pre-trainedLanguage Models and Component-based Synthesis](https://mlsanity.com/paper/2109.02445/) (2021)
- [Data Extraction via Semantic Regular Expression Synthesis](https://mlsanity.com/paper/2305.10401/) (2023)
- [FlashProfile: A Framework for Synthesizing Data Profiles](https://mlsanity.com/paper/1709.05725/) (2017)
- [Automatic Weighted Matching Rectifying Rule Discovery for Data Repairing](https://mlsanity.com/paper/1909.09807/) (2019)

## Built on by (most cited first, 6 of 6)

- [Data Analysis in the Era of Generative AI](https://mlsanity.com/paper/2409.18475/) (2024)
- [Diffusion is a code repair operator and generator](https://mlsanity.com/paper/2508.11110/) (2025)
- [Towards Cross-Modal Error Detection with Tables and Images](https://mlsanity.com/paper/2510.12383/) (2025)
- [Tabularis Formatus: Predictive Formatting for Tables](https://mlsanity.com/paper/2508.11121/) (2025)
- [ArtiFact: A Large-Scale Multi-Modal Cultural Heritage Dataset](https://mlsanity.com/paper/2606.09648/) (2026)
- [TEAR: Table Extraction with Attribute Recommendation from Texts via Large Language Models](https://mlsanity.com/paper/2609.15205/) (2026)

## Abstract (arXiv)

String data is common in real-world datasets: 67.6% of values in a sample of 1.8 million real Excel spreadsheets from the web were represented as text. Systems that successfully clean such string data can have a significant impact on real users. While prior work has explored errors in string data, proposed approaches have often been limited to error detection or require that the user provide annotations, examples, or constraints to fix the errors. Furthermore, these systems have focused independently on syntactic errors or semantic errors in strings, but ignore that strings often contain both syntactic and semantic substrings. We introduce DataVinci, a fully unsupervised string data error detection and repair system. DataVinci learns regular-expression-based patterns that cover a majority of values in a column and reports values that do not satisfy such patterns as data errors. DataVinci can automatically derive edits to the data error based on the majority patterns and constraints learned over other columns without the need for further user interaction. To handle strings with both syntactic and semantic substrings, DataVinci uses an LLM to abstract (and re-concretize) portions of strings that are semantic prior to learning majority patterns and deriving edits. Because not all data can result in majority patterns, DataVinci leverages execution information from an existing program (which reads the target data) to identify and correct data repairs that would not otherwise be identified. DataVinci outperforms 7 baselines on both error detection and repair when evaluated on 4 existing and new benchmarks.