no code implementations • 24 Apr 2023 • Timothée Fronteau, Arnaud Paran, Aymen Shabou
Adversarial attacks and defenses have gained increasing interest on computer vision systems in recent years, but as of today, most investigations are limited to images.
1 code implementation • 24 Apr 2023 • Mohamed Dhouib, Ghassen Bettaieb, Aymen Shabou
Information Extraction from visually rich documents is a challenging task that has gained a lot of attention in recent years due to its importance in several document-control based applications and its widespread commercial value.
Optical Character Recognition Optical Character Recognition (OCR)
no code implementations • 6 Jan 2021 • Sofian Chaybouti, Achraf Saghe, Aymen Shabou
This new model, that we call EfficientQA, takes advantage from the pair of sequences kind of input of BERT-based models to build meaningful dense representations of candidate answers.
Ranked #124 on Question Answering on SQuAD1.1
Extractive Question-Answering Natural Language Understanding +1
no code implementations • 17 Dec 2020 • Sofian Chaybouti, Achraf Saghe, Aymen Shabou
In this paper, we introduce MIX : a multi-task deep learning approach to solve Open-Domain Question Answering.
no code implementations • 5 Oct 2020 • Mohamed Kerroumi, Othmane Sayem, Aymen Shabou
We introduce a novel approach for scanned document representation to perform field extraction.
Ranked #1 on Document Layout Analysis on RVL-CDIP