PORTAL: Scalable Tabular Foundation Models via Content-Specific Tokenization
This provides a practical solution for researchers and practitioners working with large-scale, messy tabular datasets, though it appears incremental as it builds on existing self-supervised learning advances.
The paper tackles the challenge of scaling self-supervised learning for tabular data by introducing PORTAL, a framework that handles diverse data modalities without preprocessing, achieving state-of-the-art performance on classification and regression tasks.
Self-supervised learning on tabular data seeks to apply advances from natural language and image domains to the diverse domain of tables. However, current techniques often struggle with integrating multi-domain data and require data cleaning or specific structural requirements, limiting the scalability of pre-training datasets. We introduce PORTAL (Pretraining One-Row-at-a-Time for All tabLes), a framework that handles various data modalities without the need for cleaning or preprocessing. This simple yet powerful approach can be effectively pre-trained on online-collected datasets and fine-tuned to match state-of-the-art methods on complex classification and regression tasks. This work offers a practical advancement in self-supervised learning for large-scale tabular data.