4 papers · 1 filter
Information Redundancy and Biases in Public Document Information Extraction Benchmarks
Seif Laatiri, Pirashanth Ratnamogan, Joel Tang +3
Advances in the Visually-rich Document Understanding (VrDU) field and particularly the Key-Information Extraction (KIE) task are marked with the emergence of efficient Transformer-…
Information Extraction from Documents: Question Answering vs Token Classification in real-world setups
Laurent Lam, Pirashanth Ratnamogan, Joël Tang +2
Research in Document Intelligence and especially in Document Key Information Extraction (DocKIE) has been mainly solved as Token Classification problem. Recent breakthroughs in bot…
Robust Domain Adaptation for Pre-trained Multilingual Neural Machine Translation Models
Mathieu Grosso, Pirashanth Ratnamogan, Alexis Mathey +2
Recent literature has demonstrated the potential of multilingual Neural Machine Translation (mNMT) models. However, the most efficient models are not well suited to specialized ind…
Information Extraction from Visually Rich Documents with Font Style Embeddings
Ismail Oussaid, William Vanhuffel, Pirashanth Ratnamogan +3
Information extraction (IE) from documents is an intensive area of research with a large set of industrial applications. Current state-of-the-art methods focus on scanned documents…