| title | layout |
|---|---|
Home |
default |
We develop large models to "understand" images, videos and natural language that fuel many intelligent applications from text completion to self-driving cars. But tabular data has long been overlooked despite its dominant presence in data-intensive systems. The majority (67%) of datasets in Google Dataset Search, for example, contain typical tabular file formats like CSV and XLS. Similarly, the top-3 most-used data management systems are all relational database management systems (RDBMS). Besides dedicated tabular file formats and database systems, tables are widely used for presenting data in documents, Wikipedia pages, papers, and presentations.
By learning latent representations from structured tabular data (possibly combined with other modalities such as free-form text), pretrained table models have shown preliminary but impressive performance for semantic parsing, question answering, table understanding, and data preparation. Considering that such tasks share fundamental properties inherent to tables, representation learning for tabular data is an important direction to explore further. These works also surfaced many open challenges such as finding effective data encodings, pretraining objectives and downstream tasks. We believe, the time has come to consider tabular data as a first-class modality for representation learning and stimulate advances in this direction.
The Table Representation Learning workshop is the first workshop in this emerging research area and has the following main goals: 1) motivating tabular data as a first-class modality for representation learning and further shaping this area, 2) show-casing impactful applications of pretrained table models and discussing future opportunities thereof, and 3) facilitating discussion and collaboration across the machine learning, natural language processing, and data management communities.
- Submission deadline: 20 September 2022
- Notifications: 20 October 2022
- Camera-ready, slides and recording upload: 3 November 2022
- Workshop: Friday 2 December 2022
We are currently seeking sponsorships to support attendees from underrepresented communities and geographies. More will follow soon.