PubTables-v2: A new large-scale dataset for full-page and multi-page table extraction
This addresses a data bottleneck for researchers in visual document understanding, enabling progress in table extraction tasks, though it is incremental as it builds on existing datasets and methods.
The authors tackled the lack of annotated data for full-page and multi-page table extraction by creating PubTables-v2, a large-scale dataset that supports challenging tasks and includes the first benchmark for multi-page table structure recognition, demonstrating its usefulness through evaluations of vision-language models and the development of POTATR.
Table extraction (TE) is a key challenge in visual document understanding. Traditional approaches detect tables first, then recognize their structure. Recently, interest has surged in developing methods, such as vision-language models (VLMs), that can extract tables directly in their full page or document context. However, progress has been difficult to demonstrate due to a lack of annotated data. To address this, we create a new large-scale dataset, PubTables-v2. PubTables-v2 supports a number of current challenging table extraction tasks. Notably, it is the first large-scale benchmark for multi-page table structure recognition. We demonstrate its usefulness by evaluating domain-specialized VLMs on these tasks and highlighting current progress. Finally, we use PubTables-v2 to create the Page-Object Table Transformer (POTATR), an image-to-graph extension of the Table Transformer to comprehensive page-level TE. Data, code, and trained models will be released.