Arctic-Extract Technical Report
This addresses the need for efficient document understanding in business applications, though it appears incremental as it builds on existing extraction methods with optimizations for deployment.
The paper tackles the problem of extracting structural data from business documents by developing Arctic-Extract, a state-of-the-art model that is deployable on resource-constrained hardware, achieving deployment on A10 GPUs with 24 GB memory and processing up to 125 A4 pages.
Arctic-Extract is a state-of-the-art model designed for extracting structural data (question answering, entities and tables) from scanned or digital-born business documents. Despite its SoTA capabilities, the model is deployable on resource-constrained hardware, weighting only 6.6 GiB, making it suitable for deployment on devices with limited resources, such as A10 GPUs with 24 GB of memory. Arctic-Extract can process up to 125 A4 pages on those GPUs, making suitable for long document processing. This paper highlights Arctic-Extract's training protocols and evaluation results, demonstrating its strong performance in document understanding.