Souhail Bakkali

Souhail Bakkali, Zuheng Ming, Mickael Coustaty et al.

Multimodal learning from document data has achieved great success lately as it allows to pre-train semantically meaningful features as a prior into a learnable downstream task. In this paper, we approach the document classification problem by learning cross-modal representations through language and vision cues, considering intra- and inter-modality relationships. Instead of merging features from different modalities into a joint representation space, the proposed method exploits high-level interactions and learns relevant semantic information from effective attention flows within and across modalities. The proposed learning objective is devised between intra- and inter-modality alignment tasks, where the similarity distribution per task is computed by contracting positive sample pairs while simultaneously contrasting negative ones in the joint representation space}. Extensive experiments on public document classification datasets demonstrate the effectiveness and the generality of our model on low-scale and large-scale datasets.

Souhail Bakkali, Sanket Biswas, Zuheng Ming et al.

Visual document understanding (VDU) has rapidly advanced with the development of powerful multi-modal language models. However, these models typically require extensive document pre-training data to learn intermediate representations and often suffer a significant performance drop in real-world online industrial settings. A primary issue is their heavy reliance on OCR engines to extract local positional information within document pages, which limits the models' ability to capture global information and hinders their generalizability, flexibility, and robustness. In this paper, we introduce GlobalDoc, a cross-modal transformer-based architecture pre-trained in a self-supervised manner using three novel pretext objective tasks. GlobalDoc improves the learning of richer semantic concepts by unifying language and visual representations, resulting in more transferable models. For proper evaluation, we also propose two novel document-level downstream VDU tasks, Few-Shot Document Image Classification (DIC) and Content-based Document Image Retrieval (DIR), designed to simulate industrial scenarios more closely. Extensive experimentation has been conducted to demonstrate GlobalDoc's effectiveness in practical settings.

Omar Hamed, Souhail Bakkali, Marie-Francine Moens et al.

This work addresses the need for a balanced approach between performance and efficiency in scalable production environments for visually-rich document understanding (VDU) tasks. Currently, there is a reliance on large document foundation models that offer advanced capabilities but come with a heavy computational burden. In this paper, we propose a multimodal early exit (EE) model design that incorporates various training strategies, exit layer types and placements. Our goal is to achieve a Pareto-optimal balance between predictive performance and efficiency for multimodal document image classification. Through a comprehensive set of experiments, we compare our approach with traditional exit policies and showcase an improved performance-efficiency trade-off. Our multimodal EE design preserves the model's predictive capabilities, enhancing both speed and latency. This is achieved through a reduction of over 20% in latency, while fully retaining the baseline accuracy. This research represents the first exploration of multimodal EE design within the VDU community, highlighting as well the effectiveness of calibration in improving confidence scores for exiting at different layers. Overall, our findings contribute to practical VDU applications by enhancing both performance and efficiency.

Musab Al-Ghadi, Joris Voerman, Souhail Bakkali et al.

The increasing use of digital technologies and mobile-based registration procedures highlights the vital role of personal identity documents (IDs) in verifying users and safeguarding sensitive information. However, the rise in counterfeit ID production poses a significant challenge, necessitating the development of reliable and efficient automated verification methods. This paper introduces IDTrust, a deep-learning framework for assessing the quality of IDs. IDTrust is a system that enhances the quality of identification documents by using a deep learning-based approach. This method eliminates the need for relying on original document patterns for quality checks and pre-processing steps for alignment. As a result, it offers significant improvements in terms of dataset applicability. By utilizing a bandpass filtering-based method, the system aims to effectively detect and differentiate ID quality. Comprehensive experiments on the MIDV-2020 and L3i-ID datasets identify optimal parameters, significantly improving discrimination performance and effectively distinguishing between original and scanned ID documents.

Anthony Mudet, Souhail Bakkali

Large-scale digitization initiatives have unlocked massive collections of historical newspapers, yet effective computational access remains hindered by OCR corruption, multilingual orthographic variation, and temporal language drift. We develop and evaluate a multilingual Retrieval-Augmented Generation pipeline specifically designed for question answering on noisy historical documents. Our approach integrates: (i) semantic query expansion and multi-query fusion using Reciprocal Rank Fusion to improve retrieval robustness against vocabulary mismatch; (ii) a carefully engineered generation prompt that enforces strict grounding in retrieved evidence and explicit abstention when evidence is insufficient; and (iii) a modular architecture enabling systematic component evaluation. We conduct comprehensive ablation studies on Named Entity Recognition and embedding model selection, demonstrating the importance of syntactic coherence in entity extraction and balanced performance-efficiency trade-offs in dense retrieval. Our end-to-end evaluation framework shows that the pipeline generates faithful answers for well-supported queries while correctly abstaining from unanswerable questions. The hybrid retrieval strategy improves recall stability, particularly benefiting from RRF's ability to smooth performance variance across query formulations. We release our code and configurations at https://anonymous.4open.science/r/RAGs-C5AE/, providing a reproducible foundation for robust historical document question answering.

Vannkinh Nom, Souhail Bakkali, Muhammad Muzzamil Luqman et al.

Text recognition is significantly influenced by font types, especially for complex scripts like Khmer. The variety of Khmer fonts, each with its unique character structure, presents challenges for optical character recognition (OCR) systems. In this study, we evaluate the impact of 19 randomly selected Khmer font types on text recognition accuracy using Pytesseract. The fonts include Angkor, Battambang, Bayon, Bokor, Chenla, Dangrek, Freehand, Kh Kompong Chhnang, Kh SN Kampongsom, Khmer, Khmer CN Stueng Songke, Khmer Savuth Pen, Metal, Moul, Odor MeanChey, Preah Vihear, Siemreap, Sithi Manuss, and iSeth First. Our comparison of OCR performance across these fonts reveals that Khmer, Odor MeanChey, Siemreap, Sithi Manuss, and Battambang achieve high accuracy, while iSeth First, Bayon, and Dangrek perform poorly. This study underscores the critical importance of font selection in optimizing Khmer text recognition and provides valuable insights for developing more robust OCR systems.

Phan Phuong Mai Chau, Souhail Bakkali, Antoine Doucet

Abstractive summarization has made significant strides in condensing and rephrasing large volumes of text into coherent summaries. However, summarizing administrative documents presents unique challenges due to domain-specific terminology, OCR-generated errors, and the scarcity of annotated datasets for model fine-tuning. Existing models often struggle to adapt to the intricate structure and specialized content of such documents. To address these limitations, we introduce DocSum, a domain-adaptive abstractive summarization framework tailored for administrative documents. Leveraging pre-training on OCR-transcribed text and fine-tuning with an innovative integration of question-answer pairs, DocSum enhances summary accuracy and relevance. This approach tackles the complexities inherent in administrative content, ensuring outputs that align with real-world business needs. To evaluate its capabilities, we define a novel downstream task setting-Document Abstractive Summarization-which reflects the practical requirements of business and organizational settings. Comprehensive experiments demonstrate DocSum's effectiveness in producing high-quality summaries, showcasing its potential to improve decision-making and operational workflows across the public and private sectors.

Vannkinh Nom, Souhail Bakkali, Muhammad Muzzamil Luqman et al.

Developing effective scene text detection and recognition models hinges on extensive training data, which can be both laborious and costly to obtain, especially for low-resourced languages. Conventional methods tailored for Latin characters often falter with non-Latin scripts due to challenges like character stacking, diacritics, and variable character widths without clear word boundaries. In this paper, we introduce the first Khmer scene-text dataset, featuring 1,544 expert-annotated images, including 997 indoor and 547 outdoor scenes. This diverse dataset includes flat text, raised text, poorly illuminated text, distant and partially obscured text. Annotations provide line-level text and polygonal bounding box coordinates for each scene. The benchmark includes baseline models for scene-text detection and recognition tasks, providing a robust starting point for future research endeavors. The KhmerST dataset is publicly accessible at https://gitlab.com/vannkinhnom123/khmerst.

Souhail Bakkali, Ziheng Ming, Mickael Coustaty et al.

In the recent past, complex deep neural networks have received huge interest in various document understanding tasks such as document image classification and document retrieval. As many document types have a distinct visual style, learning only visual features with deep CNNs to classify document images have encountered the problem of low inter-class discrimination, and high intra-class structural variations between its categories. In parallel, text-level understanding jointly learned with the corresponding visual properties within a given document image has considerably improved the classification performance in terms of accuracy. In this paper, we design a self-attention-based fusion module that serves as a block in our ensemble trainable network. It allows to simultaneously learn the discriminant features of image and text modalities throughout the training stage. Besides, we encourage mutual learning by transferring the positive knowledge between image and text modalities during the training stage. This constraint is realized by adding a truncated-Kullback-Leibler divergence loss Tr-KLD-Reg as a new regularization term, to the conventional supervised setting. To the best of our knowledge, this is the first time to leverage a mutual learning approach along with a self-attention-based fusion module to perform document image classification. The experimental results illustrate the effectiveness of our approach in terms of accuracy for the single-modal and multi-modal modalities. Thus, the proposed ensemble self-attention-based mutual learning model outperforms the state-of-the-art classification results based on the benchmark RVL-CDIP and Tobacco-3482 datasets.

Souhail Bakkali, Zuheng Ming, Muhammad Muzzamil Luqman et al.

Benefiting from the advance of deep convolutional neural network approaches (CNNs), many face detection algorithms have achieved state-of-the-art performance in terms of accuracy and very high speed in unconstrained applications. However, due to the lack of public datasets and due to the variation of the orientation of face images, the complex background and lighting, defocus and the varying illumination of camera captured images, face detection on identity documents under unconstrained environments has not been sufficiently studied. To address this problem more efficiently, we survey three state-of-the-art face detection methods based on general images, i.e. Cascade-CNN, MTCNN and PCN, for face detection in camera captured images of identity documents, given different image quality assessments. For that, The MIDV-500 dataset, which is the largest and most challenging dataset for identity documents, is used to evaluate the three methods. The evaluation results show the performance and the limitations of the current methods for face detection on identity documents under the wild complex environments. These results show that the face detection task in camera captured images of identity documents is challenging, providing a space to improve in the future works.