SALT: Sales Autocompletion Linked Business Tables Dataset
Tassilo Klein, Clemens Biehl, Margarida Costa, Andre Sres, Jonas Kolk, Johannes Hoffart, SAP SE · 2025
Read SALT: Sales Autocompletion Linked Business Tables Dataset by Tassilo Klein, Clemens Biehl, Margarida Costa, Andre Sres, Jonas Kolk, Johannes Hoffart, SAP SE (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-03413.txt.
Description
Tassilo Klein and co-authors introduce SALT, a 2025 dataset of linked business tables drawn from a real enterprise planning system. Foundation models shine on text and images, but linked tables with foreign keys trip them up. Business data is exactly that shape, yet research sets from real firms are rare.
SALT curates sales tables whose rows point across tables by key, as in live company software. The set is built to train and test table representation learning, the craft of turning rows into model-ready vectors. By opening authentic enterprise data to researchers, the team hopes to sharpen models for real business tasks. Autocompletion of sales records is the namesake use case. The dataset fills a gap between toy tables and the messy data firms actually keep.
Catalog record
| CREATOR | Tassilo Klein, Clemens Biehl, Margarida Costa, Andre Sres, Jonas Kolk, Johannes Hoffart, SAP SE |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 26 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2501-03413