Wissam Siblini

Van Bach Nguyen, Kanishka Ghosh Dastidar, Michael Granitzer et al.

Fraud detection systems (FDS) mainly perform two tasks: (i) real-time detection while the payment is being processed and (ii) posterior detection to block the card retrospectively and avoid further frauds. Since human verification is often necessary and the payment processing time is limited, the second task manages the largest volume of transactions. In the literature, fraud detection challenges and algorithms performance are widely studied but the very formulation of the problem is never disrupted: it aims at predicting if a transaction is fraudulent based on its characteristics and the past transactions of the cardholder. Yet, in posterior detection, verification often takes days, so new payments on the card become available before a decision is taken. This is our motivation to propose a new paradigm: posterior fraud detection with "future" information. We start by providing evidence of the on-time availability of subsequent transactions, usable as extra context to improve detection. We then design a Bidirectional LSTM to make use of these transactions. On a real-world dataset with over 30 million transactions, it achieves higher performance than a regular LSTM, which is the state-of-the-art classifier for fraud detection that only uses the past context. We also introduce new metrics to show that the proposal catches more frauds, more compromised cards, and based on their earliest frauds. We believe that future works on this new paradigm will have a significant impact on the detection of compromised cards.

Kenneth Enevoldsen, Isaac Chung, Imene Kerboua et al. · cambridge, meta-ai

Text embeddings are typically evaluated on a limited set of tasks, which are constrained by language, domain, and task diversity. To address these limitations and provide a more comprehensive evaluation, we introduce the Massive Multilingual Text Embedding Benchmark (MMTEB) - a large-scale, community-driven expansion of MTEB, covering over 500 quality-controlled evaluation tasks across 250+ languages. MMTEB includes a diverse set of challenging, novel tasks such as instruction following, long-document retrieval, and code retrieval, representing the largest multilingual collection of evaluation tasks for embedding models to date. Using this collection, we develop several highly multilingual benchmarks, which we use to evaluate a representative set of models. We find that while large language models (LLMs) with billions of parameters can achieve state-of-the-art performance on certain language subsets and task categories, the best-performing publicly available model is multilingual-e5-large-instruct with only 560 million parameters. To facilitate accessibility and reduce computational cost, we introduce a novel downsampling method based on inter-task correlation, ensuring a diverse selection while preserving relative model rankings. Furthermore, we optimize tasks such as retrieval by sampling hard negatives, creating smaller but effective splits. These optimizations allow us to introduce benchmarks that drastically reduce computational demands. For instance, our newly introduced zero-shot English benchmark maintains a ranking order similar to the full-scale version but at a fraction of the computational cost.

Wissam Siblini, Guillaume Coter, Rémy Fabry et al.

The dark face of digital commerce generalization is the increase of fraud attempts. To prevent any type of attacks, state-of-the-art fraud detection systems are now embedding Machine Learning (ML) modules. The conception of such modules is only communicated at the level of research and papers mostly focus on results for isolated benchmark datasets and metrics. But research is only a part of the journey, preceded by the right formulation of the business problem and collection of data, and followed by a practical integration. In this paper, we give a wider vision of the process, on a case study of transfer learning for fraud detection, from business to research, and back to business.

Wissam Siblini, Mohamed Challal, Charlotte Pasqual

Open Domain Question Answering (ODQA) on a large-scale corpus of documents (e.g. Wikipedia) is a key challenge in computer science. Although transformer-based language models such as Bert have shown on SQuAD the ability to surpass humans for extracting answers in small passages of text, they suffer from their high complexity when faced to a much larger search space. The most common way to tackle this problem is to add a preliminary Information Retrieval step to heavily filter the corpus and only keep the relevant passages. In this paper, we propose a more direct and complementary solution which consists in applying a generic change in the architecture of transformer-based models to delay the attention between subparts of the input and allow a more efficient management of computations. The resulting variants are competitive with the original models on the extractive task and allow, on the ODQA setting, a significant speedup and even a performance improvement in many cases.

Wissam Siblini, Charlotte Pasqual, Axel Lavielle et al.

Recent advances with language models (e.g. BERT, XLNet, ...), have allowed surpassing human performance on complex NLP tasks such as Reading Comprehension. However, labeled datasets for training are available mostly in English which makes it difficult to acknowledge progress in other languages. Fortunately, models are now pre-trained on unlabeled data from hundreds of languages and exhibit interesting transfer abilities from one language to another. In this paper, we show that multilingual BERT is naturally capable of zero-shot transfer for an extractive Question Answering task (eQA) from English to other languages. More specifically, it outperforms the best previously known baseline for transfer to Japanese and French. Moreover, using a recently published large eQA French dataset, we are able to further show that (1) zero-shot transfer provides results really close to a direct training on the target language and (2) combination of transfer and training on target is the best option overall. We finally present a practical application: a multilingual conversational agent called Kate which answers to HR-related questions in several languages directly from the content of intranet pages.

Wissam Siblini, Jordan Fréry, Liyun He-Guelton et al.

Machine learning models deployed in real-world applications are often evaluated with precision-based metrics such as F1-score or AUC-PR (Area Under the Curve of Precision Recall). Heavily dependent on the class prior, such metrics make it difficult to interpret the variation of a model's performance over different subpopulations/subperiods in a dataset. In this paper, we propose a way to calibrate the metrics so that they can be made invariant to the prior. We conduct a large number of experiments on balanced and imbalanced data to assess the behavior of calibrated metrics and show that they improve interpretability and provide a better control over what is really measured. We describe specific real-world use-cases where calibration is beneficial such as, for instance, model monitoring in production, reporting, or fairness evaluation.

Wissam Siblini, Frank Meyer, Pascale Kuntz

This paper presents a new interactive opinion mining tool that helps users to classify large sets of short texts originated from Web opinion polls, technical forums or Twitter. From a manual multi-label pre-classification of a very limited text subset, a learning algorithm predicts the labels of the remaining texts of the corpus and the texts most likely associated to a selected label. Using a fast matrix factorization, the algorithm is able to handle large corpora and is well-adapted to interactivity by integrating the corrections proposed by the users on the fly. Experimental results on classical datasets of various sizes and feedbacks of users from marketing services of the telecommunication company Orange confirm the quality of the obtained results.