Nadir Durrani

• PhD
• Researcher at Qatar Computing Research Institute

We present Fanar, a platform for Arabic-centric multimodal generative AI systems, that supports language, speech and image generation tasks. At the heart of Fanar are Fanar Star and Fanar Prime, two highly capable Arabic Large Language Models (LLMs) that are best in the class on well established benchmarks for similar sized models. Fanar Star is a...

Arabic, with its rich diversity of dialects, remains significantly underrepresented in Large Language Models, particularly in dialectal variations. We address this gap by introducing seven synthetic datasets in dialects alongside Modern Standard Arabic (MSA), created using Machine Translation (MT) combined with human post-editing. We present AraDiC...

Recent advancements in Large Language Models (LLMs) have significantly influenced the landscape of language and speech research. Despite this progress, these models lack specific benchmarking against state-of-the-art (SOTA) models tailored to particular languages and tasks. LAraBench addresses this gap for Ara-bic Natural Language Processing (NLP)...

Despite their remarkable ability to capture linguistic nuances across diverse languages, questions persist regarding the degree of alignment between languages in multilingual embeddings. Drawing inspiration from research on high-dimensional representations in neural language models, we employ clustering to uncover latent concepts within multilingua...

Pre-trained language models (pLMs) learn intricate patterns and contextual dependencies via unsupervised learning on vast text data, driving breakthroughs across NLP tasks. Despite these achievements, these models remain black boxes, necessitating research into understanding their decision-making processes. Recent studies explore representation ana...

The recent development and success of Large Language Models (LLMs) necessitate an evaluation of their performance across diverse NLP tasks in different languages. Although several frameworks have been developed and made publicly available, their customization capabilities for specific tasks and datasets are often complex for different users. In thi...

The opacity of deep neural networks remains a challenge in deploying solutions where explanation is as important as precision. We present ConceptX, a human-in-the-loop framework for interpreting and annotating latent representational space in pre-trained Language Models (pLMs). We use an unsupervised method to discover concepts learned in these mod...

Neuron analysis provides insights into how knowledge is structured in representations and discovers the role of neurons in the network. In addition to developing an understanding of our models, neuron analysis enables various applications such as debiasing, domain adaptation and architectural search. We present NeuroX, a comprehensive open-source t...

With large Foundation Models (FMs), language technologies (AI in general) are entering a new paradigm: eliminating the need for developing large-scale task-specific datasets and supporting a variety of tasks through set-ups ranging from zero-shot to few-shot learning. However, understanding FMs capabilities requires a systematic benchmarking effort...

Work done to uncover the knowledge encoded within pre-trained language models, rely on annotated corpora or human-in-the-loop methods. However, these approaches are limited in terms of scalability and the scope of interpretation. We propose using a large language model, ChatGPT, as an annotator to enable fine-grained interpretation analysis of pre-...

Neuron Interpretation has gained traction in the field of interpretability, and have provided fine-grained insights into what a model learns and how language knowledge is distributed amongst its different components. However, the lack of evaluation benchmark and metrics have led to siloed progress within these various methods, with very little work...

The proliferation of Deep Neural Networks in various domains has seen an increased need for interpretability of these models. Preliminary work done along this line, and papers that surveyed such, are focused on high-level representation analysis. However, a recent branch of work has concentrated on interpretability at a more granular level of analy...

The opacity of deep neural networks remains a challenge in deploying solutions where explanation is as important as precision. We present ConceptX, a human-in-the-loop framework for interpreting and annotating latent representational space in pre-trained Language Models (pLMs). We use an unsupervised method to discover concepts learned in these mod...

We study the evolution of latent space in fine-tuned NLP models. Different from the commonly used probing-framework, we opt for an unsupervised method to analyze representations. More specifically, we discover latent concepts in the representational space using hierarchical clustering. We then use an alignment function to gauge the similarity betwe...

Arabic is a Semitic language which is widely spoken with many dialects. Given the success of pre-trained language models, many transformer models trained on Arabic and its dialects have surfaced. While there have been an extrinsic evaluation of these models with respect to downstream NLP tasks, no work has been carried out to analyze and compare th...

Transformer-based NLP models are trained using hundreds of millions or even billions of parameters, limiting their applicability in computationally constrained environments. While the number of parameters generally correlates with performance, it is not clear whether the entire network is required for a downstream task. Motivated by the recent work...

We propose a novel framework ConceptX, to analyze how latent concepts are encoded in representations learned within pre-trained language models. It uses clustering to discover the encoded concepts and explains them by aligning with a large set of human-defined concepts. Our analysis on seven transformer language models reveal interesting insights:...

While a lot of work has been done in understanding representations learned within deep NLP models and what knowledge they capture, little attention has been paid towards individual neurons. We present a technique called as Linguistic Correlation Analysis to extract salient neurons in the model, with respect to any extrinsic property - with the goal...

NatiQ is end-to-end text-to-speech system for Arabic. Our speech synthesizer uses an encoder-decoder architecture with attention. We used both tacotron-based models (tacotron-1 and tacotron-2) and the faster transformer model for generating mel-spectrograms from characters. We concatenated Tacotron1 with the WaveRNN vocoder, Tacotron2 with the Wave...

A large number of studies that analyze deep neural network models and their ability to encode various linguistic and non-linguistic concepts provide an interpretation of the inner mechanics of these models. The scope of the analyses is limited to pre-defined concepts that reinforce the traditional linguistic knowledge and do not reflect on how nove...

A large number of studies that analyze deep neural network models and their ability to encode various linguistic and non-linguistic concepts provide an interpretation of the inner mechanics of these models. The scope of the analyses is limited to pre-defined concepts that reinforce the traditional linguistic knowledge and do not reflect on how nove...

Arabic is a Semitic language which is widely spoken with many dialects. Given the success of pre-trained language models, many transformer models trained on Arabic and its dialects have surfaced. While these models have been compared with respect to downstream NLP tasks, no evaluation has been carried out to directly compare the internal representa...

The proliferation of deep neural networks in various domains has seen an increased need for interpretability of these methods. A plethora of research has been carried out to analyze and understand components of the deep neural network models. Preliminary work done along these lines and papers that surveyed such, were focused on a more high-level re...

End-to-end deep neural network architectures have pushed the state-of-the-art in speech technologies, as well as in other spheres of Artificial Intelligence, subsequently leading researchers to train more complex and deeper models. These improvements came at the cost of transparency. Deep neural networks are innately opaque and difficult to interpr...

Transfer learning from pre-trained neural language models towards downstream tasks has been a predominant theme in NLP recently. Several researchers have shown that deep NLP models learn non-trivial amount of linguistic knowledge, captured at different layers of the model. We investigate how fine-tuning towards downstream NLP tasks impacts the lear...

With the outbreak of the COVID-19 pandemic, people turned to social media to read and to share timely information including statistics, warnings, advice, and inspirational stories. Unfortunately, alongside all this useful information, there was also a new blending of medical and political misinformation and disinformation, which gave rise to the fi...

This paper is a write-up for the tutorial on "Fine-grained Interpretation and Causation Analysis in Deep NLP Models" that we are presenting at NAACL 2021. We present and discuss the research work on interpreting fine-grained components of a model from two perspectives, i) fine-grained interpretation, ii) causation analysis. The former introduces me...

Post-processing of static embedding has beenshown to improve their performance on both lexical and sequence-level tasks. However, post-processing for contextualized embeddings is an under-studied problem. In this work, we question the usefulness of post-processing for contextualized embeddings obtained from different layers of pre-trained language...

While a lot of analysis has been carried to demonstrate linguistic knowledge captured by the representations learned within deep NLP models, very little attention has been paid towards individual neurons.We carry outa neuron-level analysis using core linguistic tasks of predicting morphology, syntax and semantics, on pre-trained language models, wi...

With the outbreak of the COVID-19 pandemic, people turned to social media to read and to share timely information including statistics, warnings, advice, and inspirational stories. Unfortunately, alongside all this useful information, there was also a new blending of medical and political misinformation and disinformation, which gave rise to the fi...

This paper investigates contextual word representation models from the lens of similarity analysis. Given a collection of trained models, we measure the similarity of their internal representations and attention. Critically, these models come from vastly different architectures. We use existing and novel similarity measures that aim to gauge the le...

Disinformation, i.e., information that is both false and means harm, thrives in social media. Most often, it is used for political purposes, e.g., to influence elections or simply to cause distrust in society. It can also target medical issues, most notably the use of vaccines. With the emergence of the COVID-19 pandemic, the political and the medi...

Disinformation, i.e., information that is both false and means harm, thrives in social media. Most often, it is used for political purposes , e.g., to influence elections or simply to cause distrust in society. It can also target medical issues, most notably the use of vaccines. With the emergence of the COVID-19 pandemic, the political and the med...

Disinformation, i.e., information that is both false and means harm, thrives in social media. Most often, it is used for political purposes , e.g., to influence elections or simply to cause distrust in society. It can also target medical issues, most notably the use of vaccines. With the emergence of the COVID-19 pandemic, the political and the med...

Large pre-trained contextual word representations have transformed the field of natural language processing, obtaining impressive results on a wide range of tasks. However, as models increase in size, computational limitations make them impractical for researchers and practitioners alike. We hypothesize that contextual representations have both int...

The ongoing neural revolution in Natural Language Processing has recently been dominated by large-scale pre-trained Transformer models, where size does matter: it has been shown that the number of parameters in such a model is typically positively correlated with its performance. Naturally, this situation has unleashed a race for ever larger models...

Despite the recent success of deep neural networks in natural language processing and other spheres of artificial intelligence, their interpretability remains a challenge. We analyze the representations learned by neural machine translation (NMT) models at various levels of granularity and evaluate their quality through relevant extrinsic propertie...

Despite the recent success of deep neural networks in natural language processing (NLP), their interpretability remains a challenge. We analyze the representations learned by neural machine translation models at various levels of granularity and evaluate their quality through relevant extrinsic properties. In particular, we seek answers to the foll...

We present a toolkit to facilitate the interpretation and understanding of neural network models. The toolkit provides several methods to identify salient neurons with respect to the model itself or an external task. A user can visualize selected neurons, ablate them to measure their effect on the model accuracy, and manipulate them to control the...

Despite the remarkable evolution of deep neural networks in natural language processing (NLP), their interpretability remains a challenge. Previous work largely focused on what these models learn at the representation level. We break this analysis down further and study individual dimensions (neurons) in the vector representation learned by end-to-...

We share the findings of the first shared task on improving robustness of Machine Translation (MT). The task provides a testbed representing challenges facing MT models deployed in the real world, and facilitates new approaches to improve models; robustness to noisy input and domain mismatch. We focus on two language pairs (English-French and Engli...

We present a toolkit to facilitate the interpretation and understanding of neural network models. The toolkit provides several methods to identify salient neurons with respect to the model itself or an external task. A user can visualize selected neurons, ablate them to measure their effect on the model accuracy , and manipulate them to control the...

Despite the remarkable evolution of deep neural networks in natural language processing (NLP), their interpretability remains a challenge. Previous work largely focused on what these models learn at the representation level. We break this analysis down further and study individual dimensions (neu-rons) in the vector representation learned by end-to...

Despite the remarkable evolution of deep neural networks in natural language processing (NLP), their interpretability remains a challenge. Previous work largely focused on what these models learn at the representation level. We break this analysis down further and study individual dimensions (neurons) in the vector representation learned by end-to-...

We present a toolkit to facilitate the interpretation and understanding of neural network models. The toolkit provides several methods to identify salient neurons with respect to the model itself or an external task. A user can visualize selected neurons, ablate them to measure their effect on the model accuracy, and manipulate them to control the...

Neural machine translation (NMT) models learn representations containing substantial linguistic information. However, it is not clear if such information is fully distributed or if some of it can be attributed to individual neurons. We develop unsupervised methods for discovering important neurons in NMT models. Our methods rely on the intuition th...

We address the problem of simultaneous translation by modifying the Neural MT decoder to operate with dynamically built encoder and attention. We propose a tunable agent which decides the best segmentation strategy for a user-defined BLEU loss and Average Proportion (AP) constraint. Our agent outperforms previously proposed Wait-if-diff and Wait-if...

We address the problem of simultaneous translation by modifying the Neural MT decoder to operate with dynamically built encoder and attention. We propose a tunable agent which decides the best segmentation strategy for a user-defined BLEU loss and Average Proportion (AP) constraint. Our agent outperforms previously proposed Wait-if-diff and Wait-if...

Bilingual sequence models improve phrase-based translation and reordering by overcoming phrasal independence assumption and handling long range reordering. However, due to data sparsity, these models often fall back to very small context sizes. This problem has been previously addressed by learning sequences over generalized representations such as...

While neural machine translation (NMT) models provide improved translation quality in an elegant, end-to-end framework, it is less clear what they learn about language. Recent work has started evaluating the quality of vector representations learned by NMT models on morphological and syntactic tasks. In this paper, we investigate the representation...

We present research towards bridging the language gap between migrant workers in Qatar and medical staff. In particular, we present the first steps towards the development of a real-world Hindi-English machine translation system for doctor-patient communication. As this is a low-resource language pair, especially for speech and for the medical doma...

In this work, we present Qatar Computing Research Institute»s live speech translation system. Our system works with both Arabic and English. It is designed using an array of modern web technologies to capture speech in real time, and transcribe and translate it using state-of-the-art Automatic Speech Recognition (ASR) and Machine Translation (MT) s...

Bilingual sequence models improve phrase-based translation and reordering by overcoming phrasal independence assumption and handling long range reordering. However, due to data sparsity, these models often fall back to very small context sizes. This problem has been previously addressed by learning sequences over generalized representations such as...

End-to-end training makes the neural machine translation (NMT) architecture simpler , yet elegant compared to traditional statistical machine translation (SMT). However, little is known about linguistic patterns of morphology, syntax and semantics learned during the training of NMT systems, and more importantly, which parts of the architecture are...

While neural machine translation (NMT) models provide improved translation quality in an elegant framework, it is less clear what they learn about language. Recent work has started evaluating the quality of vector representations learned by NMT models on morphological and syntactic tasks. In this paper, we investigate the representations learned at...

Word segmentation plays a pivotal role in improving any Arabic NLP application. Therefore, a lot of research has been spent in improving its accuracy. Off-the-shelf tools, however, are: i) complicated to use and ii) domain/dialect dependent. We explore three language-independent alternatives to morphological segmentation using: i) data-driven sub-w...

In this paper, we explore alternative ways to train a neural machine translation system in a multi-domain scenario. We investigate data concatenation (with fine tuning), model stacking (multi-level fine tuning), data selection and weighted ensemble. We evaluate these methods based on three criteria: i) translation quality, ii) training time, and ii...

Neural machine translation (MT) models obtain state-of-the-art performance while maintaining a simple, end-to-end architecture. However, little is known about what these models learn about source and target languages during the training process. In this work, we analyze the representations learned by neural MT models at various levels of granularit...

We present the first prototype of the SUMMA Platform: an integrated platform for multilingual media monitoring. The platform contains a rich suite of low-level and high-level natural language processing technologies: automatic speech recognition of broadcast media, machine translation, automated tagging and classification of named entities, semanti...

We present QCRI's Arabic-to-English speech translation system. It features modern web technologies to capture live audio, and broadcasts Arabic transcriptions and English translations simultaneously. Our Kaldi-based ASR system uses the Time Delay Neural Network architecture , while our Machine Translation (MT) system uses both phrase-based and neur...

In this study, we present an analysis regarding the performance of the state-of-art Phrase-based Statistical Machine Translation (SMT) on multiple Indian languages. We report baseline systems on several language pairs. The motivation of this study is to promote the development of SMT and linguistic resources for these language pairs, as the current...

This paper describes QCRI's machine translation systems for the IWSLT 2016 evaluation campaign. We participated in the Arabic->English and English->Arabic tracks. We built both Phrase-based and Neural machine translation models, in an effort to probe whether the newly emerged NMT framework surpasses the traditional phrase-based systems in Arabic-En...

We explore neural joint models for the task of domain adaptation in machine translation in two ways: (i) we apply state-of-the-art domain adaptation techniques, such as mixture modelling and data selection using the recently proposed Neural Network Joint Model (NNJM) (Devlin et al., 2014); (ii) we propose two novel approaches to perform adaptation...

We present a novel fusion model for domain adaptation in Statistical Machine Translation. Our model is based on the joint source-target neural network (Devlin et al., 2014), and is learned by fusing in-and out-domain models. The adaptation is performed by backpropagating errors from the output layer to the word embedding layer of each model, subseq...

This paper describes QCRI's machine translation systems for the IWSLT 2016 evaluation campaign. We participated in the Arabic→English and English→Arabic tracks. We built both Phrase-based and Neural machine translation models, in an effort to probe whether the newly emerged NMT framework surpasses the traditional phrase-based systems in Arabic-Engl...

We present research towards bridging the language gap between migrant workers in Qatar and medical staff. In particular, we present the first steps towards the development of a real-world Hindi-English machine translation system for doctor-patient communication. As this is a low-resource language pair, especially for speech and for the medical doma...

The paper describes the Egyptian Arabic-to-English statistical machine translation (SMT) system that the QCRI-Columbia-NYUAD (QCN) group submitted to the NIST OpenMT'2015 competition. The competition focused on informal dialectal Arabic, as used in SMS, chat, and speech. Thus, our efforts focused on processing and standardizing Arabic, e.g., using...

In this paper, we present Farasa, a fast and accurate Arabic segmenter. Our approach is based on SVM-rank using linear kernels. We measure the performance of the seg-menter in terms of accuracy and efficiency, in two NLP tasks, namely Machine Translation (MT) and Information Retrieval (IR). Farasa outperforms or is at par with the state-of-the-art...

Poorly translated text is often disfluent and difficult to read. In contrast, well-formed translations require less time to process. In this paper, we model the differences in reading patterns of Machine Translation (MT) evalua-tors using novel features extracted from their gaze data, and we learn to predict the quality scores given by those evalua...

We present iAppraise: an open-source framework that enables the use of eye-tracking for MT evaluation. It connects Appraise, an open-source toolkit for MT evaluation, to a low-cost eye-tracking device, to make its usage accessible to a broader audience. It also provides a set of tools for extracting and exploiting gaze data, which facilitate eye-tr...

We present research towards bridging the language gap between migrant workers in Qatar and medical staff. In particular, we present the first steps towards the development of a real-world Hindi-English machine translation system for doctor-patient communication. As this is a low-resource language pair, especially for speech and for the medical doma...

Joint models have recently shown to improve the state-of-the-art in machine translation (MT). We apply EM-based mixture modeling and data selection techniques using two joint models, namely the Operation Sequence Model or OSM — an ngram-based translation and reordering model, and the Neural Network Joint Model or NNJM — a continuous space translati...

We present novel models for domain adaptation based on the neural network joint model (NNJM). Our models maximize the cross entropy by regularizing the loss function with respect to in-domain model. Domain adaptation is carried out by assigning higher weight to out-domain sequences that are similar to the in-domain data. In our alternative model we...