Large language models (LLMs) developed, deployed, and researched across the GCC — including Falcon (TII), Jais (MBZUAI/Inception), AceGPT (KAUST), ALLaM (SDAIA), and Fanar (QCRI).
Dr. Mikhail Burtsev of the London Institute presented research on GENA-LM, a suite of transformer-based DNA language models. The talk addressed the challenge of scaling transformers for genomic sequences, proposing recurrent memory augmentation to handle long input sequences efficiently. This approach improves language modeling performance and holds promise for memory-intensive applications in bioinformatics. Why it matters: This research can significantly advance AI's capabilities in genomics by enabling the processing of much larger DNA sequences, with potential breakthroughs in understanding and treating diseases.
A talk will present two projects related to the use of NLP for estimating a client’s depression severity and well-being. The first project examines emotional coherence between the subjective experience of emotions and emotion expression in therapy using transformer-based emotion recognition models. The second project proposes a semantic pipeline to study depression severity in individuals based on their social media posts by exploring different aggregation methods to answer one of four Beck Depression Inventory (BDI) options per symptom. Why it matters: This research explores how NLP techniques can be applied to mental health assessment, potentially offering new tools for diagnosis and treatment monitoring.
Zeerak Talat, an independent scholar, gave a talk at MBZUAI on automated content moderation and the impacts of machine learning on society. Talat's research considers how machine learning interacts with and impacts societies through content moderation technologies, drawing from NLP, privacy preserving machine learning, science and technology studies, decolonize studies, and media studies. The talk highlighted research areas that can afford productive directions for the meeting between machine learning and society. Why it matters: The talk contributes to the discussion of ethical AI development and deployment in the region, particularly regarding content moderation and its societal impacts.
The article discusses parameter-efficient fine-tuning methods for large NLP models, highlighting their importance due to the increasing size and computational demands of state-of-the-art language models. It provides an overview of these methods, presenting them in a unified view to emphasize their similarities and differences. Indraneil, a PhD candidate at TU Darmstadt's UKP Lab, is researching parameter-efficient fine-tuning, sparsity, and conditional computation methods to improve LLM performance in multilingual, multi-task settings. Why it matters: Efficient fine-tuning techniques are crucial for democratizing access to and accelerating the deployment of large language models in the region and beyond.
A new content improvement system has been developed to address issues of randomness and incorrectness in text generated by deep learning models like GPT-3. The system uses text mining to identify correct sentences and employs syntactic/semantic generalization to substitute problematic elements. The system can substantially improve the factual correctness and meaningfulness of raw content. Why it matters: Improving the quality of automatically generated content is crucial for ensuring reliability and trustworthiness across various AI applications.
NYU and NYU Abu Dhabi researchers are working on user-centric gender rewriting in NLP, especially for Arabic. They are building an Arabic Parallel Gender Corpus and developing models for gender rewriting tasks. The work aims to address representational harms caused by NLP systems that don't account for user preferences regarding grammatical gender. Why it matters: This research promotes fairness and inclusivity in Arabic NLP by enabling systems to generate gender-specific outputs based on user preferences, mitigating biases present in training data.
Project LITMUS explores predicting cross-lingual transfer accuracy in multilingual language models, even without test data in target languages. The goal is to estimate model performance in low-resource languages and optimize training data for desired cross-lingual performance. This research aims to identify factors influencing cross-lingual transfer, contributing to linguistically fair MMLMs. Why it matters: Improving cross-lingual transfer is vital for creating more equitable and effective multilingual AI systems, especially for languages with limited resources.
This talk explores modern machine learning through high-dimensional statistics, using random matrix theory to analyze learning models. The speaker, Denny Wu from University of Toronto and the Vector Institute, presents two examples: hyperparameter selection in overparameterized models and gradient-based representation learning in neural networks. The analysis reveals insights such as the possibility of negative optimal ridge penalty and the advantages of feature learning over random features. Why it matters: This research provides a deeper theoretical understanding of deep learning phenomena, with potential implications for optimizing training and improving model performance in the region.
A talk at MBZUAI discussed federated learning, a distributed machine learning approach training models over devices while keeping data localized. The presentation covered a straggler-resilient federated learning scheme using adaptive node participation to tackle system heterogeneity. It also presented a robust optimization formulation for addressing data heterogeneity and a new algorithm for personalizing learned models. Why it matters: Federated learning is crucial for AI applications involving decentralized data sources, and research on improving its robustness and personalization is essential for real-world deployment in the region.
Thamar Solorio from the University of Houston will discuss machine learning approaches for spontaneous human language processing. The talk will cover adapting multilingual transformers to code-switching data and using data augmentation for domain adaptation in sequence labeling tasks. Solorio will also provide an overview of other research projects at the RiTUAL lab, focusing on the scarcity of labeled data. Why it matters: This presentation addresses key challenges in Arabic NLP related to data scarcity, which is a persistent obstacle in developing effective AI applications for the region.
Fudan University's Zhongyu Wei presented research on social simulation driven by LLMs, covering individual and large-scale social movement simulation. Wei directs the Data Intelligence and Social Computing Lab (Fudan DISC) and has published extensively on multimodal large models and social computing. His work includes the Volcano multimodal model, DISC-MedLLM, and ElectionSim. Why it matters: Using LLMs for social simulation could provide new tools for understanding and potentially predicting social dynamics in the Arab world.
A Caltech researcher presented at MBZUAI on memory representation and retrieval, contrasting AI and neuroscience approaches. Current AI retrieval systems like RAG retrieve via fine-tuning and embedding similarity, while the presenter argued for exploring retrieval via combinatorial object identity or spatial proximity. The research explores circuit-level retrieval via domain fine-tuned LLMs and distributed memory for image retrieval using semantic similarity. Why it matters: The work suggests structured databases and retrieval-focused training can allow smaller models to outperform larger general-purpose models, offering efficiency gains for AI development in the region.
A KAUST alumnus presented research on using large language models for complex disease modeling and drug discovery. LLMs were trained on insurance claims of 123 million US people to model diseases and predict genetic parameters. Protein language models were developed to discover remote homologs and functional biomolecules, while RNA language models were used for RNA structure prediction and reverse design. Why it matters: This work highlights the potential of LLMs to accelerate computational biology research and drug development, with a KAUST connection.
Dr. Zeke Xie from HKUST(GZ) presented research on noise initialization and sampling strategies for diffusion models. The talk covered golden noise for text-to-image models, zigzag diffusion sampling, smooth initializations for video diffusion, and leveraging image diffusion for video synthesis. Xie leads the xLeaF Lab, focusing on optimization, inference, and generative AI, with previous experience at Baidu Research. Why it matters: The work addresses core challenges in improving the quality and diversity of generated content from diffusion models, a key area of advancement for AI applications in the region.
Maha Elgarf from NYU Abu Dhabi presented research on using social robots to stimulate creativity in children through subconscious mimicry, leveraging the 'chameleon effect'. The research involved a series of studies where children engaged in storytelling with a social robot, and their creativity was assessed. Elgarf also discussed using Large Language Models (LLMs) in education and challenges in the field. Why it matters: This explores innovative applications of social robotics and AI in education within the UAE, potentially enhancing children's learning and creativity.
Jan Buchmann from TU Darmstadt presented research on NLP for long, structured documents at MBZUAI. The research addresses gaps in using document structure and improving the verifiability of LM responses. Experiments showed that models learn to represent document structure during pre-training, and larger models can cite sources well. Why it matters: This research contributes to making NLP more effective for complex documents like scientific articles and legal texts, which is crucial for information accessibility.
Hassan Sajjad from Dalhousie University presented research on exploring the latent space of AI models to assess their safety and trustworthiness. He discussed use cases where analyzing latent space helps understand the robustness-generalization tradeoff in adversarial training and evaluate language comprehension. Sajjad's work aims to build better AI models and increase trust in their capabilities by looking at model internals. Why it matters: Intrinsic evaluation of model internals will become important to improving AI safety and robustness.
Pedro J. Moreno, former head of ASR R&D at Google, presented a talk at MBZUAI on the past, present, and future of speech technologies. The talk covered the evolution of speech tech, his career contributions including work on Google Voice search, and the impact of LLMs on speech science. He also discussed the interplay between foundational and applied research and preparing the next generation of scientists. Why it matters: The talk provides insights into the trajectory of speech technologies from a leading researcher, highlighting future directions and the ethical considerations surrounding AI's impact on society.
MBZUAI's Dr. Artem Shelmanov is working on uncertainty quantification (UQ) methods for generative LLMs to detect unreliable generations. He aims to address the issue of LLMs fabricating facts, often called "hallucinating", without clear indicators of veracity. He systemizes existing UQ efforts, discusses caveats, and suggests novel techniques for safer LLM use. Why it matters: Improving the reliability of LLMs is crucial for responsible AI deployment in the region, especially in sensitive applications.
Dr. Jindong Wang from Microsoft Research Asia gave a talk at MBZUAI about the limitations of large foundation models, including adapting to real-world unpredictability and security concerns. He also discussed the need for interdisciplinary collaboration to evaluate the benefits and risks of these models. Dr. Wang shared his research and insights on how to harness the power of large foundation models while addressing their constraints and fostering responsible AI integration. Why it matters: This highlights MBZUAI's role in hosting discussions about responsible AI development and the challenges of deploying foundation models.
Iryna Gurevych from TU Darmstadt presented research on using large language models for real-world fact-checking, focusing on dismantling misleading narratives from misinterpreted scientific publications and detecting misinformation via visual content. The research aims to explain why a false claim was believed, why it is false, and why the alternative is correct. Why it matters: Addressing misinformation, especially when supported by seemingly credible sources, is critical for public health, conflict resolution, and maintaining trust in institutions in the Middle East and globally.
Michael Kampffmeyer from UiT The Arctic University of Norway presented a talk at MBZUAI on representation learning for deep clustering and few-shot learning. The talk covered deep clustering in multi-view settings and the influence of geometrical representation properties on few-shot classification performance. He specifically discussed embedding representations on the hypersphere and its connection to the hubness phenomenon. Why it matters: This highlights MBZUAI's role in hosting discussions on advanced machine learning topics like few-shot learning, which are crucial for addressing data scarcity challenges in the region and beyond.
Adel Bibi, a KAUST alumnus and researcher at the University of Oxford, presented his research on AI safety, covering robustness, alignment, and fairness of LLMs. The research addresses challenges in AI systems, alignment issues, and fairness across languages in common tokenizers. Bibi's work includes instruction prefix tuning and its theoretical limitations towards alignment. Why it matters: This research from a leading researcher highlights the importance of addressing safety concerns in LLMs, particularly regarding alignment and fairness in the Arabic language.
The article discusses research on fine-tuning text-to-image diffusion models, including reward function training, online reinforcement learning (RL) fine-tuning, and addressing reward over-optimization. A Text-Image Alignment Assessment (TIA2) benchmark is introduced to study reward over-optimization. TextNorm, a method for confidence calibration in reward models, is presented to reduce over-optimization risks. Why it matters: Improving the alignment and fidelity of text-to-image models is crucial for generating high-quality content, and addressing over-optimization enhances the reliability of these models in creative applications.
Xiuying Chen from KAUST presented her work on improving the trustworthiness of AI-generated text, focusing on accuracy and robustness. Her research analyzes causes of hallucination in language models related to semantic understanding and neglect of input knowledge, and proposes solutions. She also demonstrated vulnerabilities of language models to noise and enhances robustness using augmentation techniques. Why it matters: Improving the reliability of AI-generated text is crucial for its deployment in sensitive domains like healthcare and scientific discovery, where accuracy is paramount.
Soufiane Hayou of the National University of Singapore presented a talk at MBZUAI on principled scaling of neural networks. The talk covered leveraging mathematical results to efficiently scale neural networks. He obtained his PhD in statistics in 2021 from Oxford. Why it matters: Understanding neural network scaling is crucial for developing more efficient and powerful AI models in the region.
This article discusses the increasing concerns about the interpretability of large deep learning models. It highlights a talk by Danish Pruthi, an Assistant Professor at the Indian Institute of Science (IISc), Bangalore, who presented a framework to quantify the value of explanations and the need for holistic model evaluation. Pruthi's talk touched on geographically representative artifacts from text-to-image models and how well conversational LLMs challenge false assumptions. Why it matters: Addressing interpretability and evaluation is crucial for building trustworthy and reliable AI systems, particularly in sensitive applications within the Middle East and globally.
MBZUAI's Dr. Fajri Koto presented research on overcoming challenges in NLP for underrepresented languages. His work includes creating multilingual datasets for Indonesian languages by engaging native speakers and finding that direct composition yields better results than translation. He also discussed vocabulary adaptation and zero-shot learning to address computational resource limitations, and emphasized the importance of datasets with local context for evaluating LLMs. Why it matters: This research addresses critical gaps in NLP for low-resource languages, providing insights and techniques to improve performance and cultural relevance in multilingual AI models within the region and globally.
Liangming Pan from UCSB presented research on building reliable generative AI agents by integrating symbolic representations with LLMs. The neuro-symbolic strategy combines the flexibility of language models with precise knowledge representation and verifiable reasoning. The work covers Logic-LM, ProgramFC, and learning from automated feedback, aiming to address LLM limitations in complex reasoning tasks. Why it matters: Improving the reliability of LLMs is crucial for high-stakes applications in finance, medicine, and law within the region and globally.
Akhil Arora from EPFL presented a framework for AI-assisted knowledge navigation, focusing on understanding and enhancing human navigation on Wikipedia. The framework includes methods for modeling navigation patterns, identifying knowledge gaps, and assessing their causal impact. He also discussed applications beyond Wikipedia, such as multimodal knowledge navigation assistants and multilingual knowledge gap mitigation. Why it matters: This research has the potential to improve information systems by making online knowledge more accessible and navigable, especially for platforms like Wikipedia that serve as critical resources for global knowledge sharing.
MBZUAI alumnus Zain Muhammed Mujahid is pursuing a Ph.D. at the University of Copenhagen, focusing on factual text generation in LLMs to combat misinformation. During his master's at MBZUAI, he researched political bias and misinformation in media outlets using LLMs, under the mentorship of Professor Preslav Nakov. His master's thesis involved assessing a media outlet's factual reporting level and political leaning on topics like immigration and economy. Why it matters: This research addresses a critical challenge in AI, enhancing the reliability of LLMs and mitigating the spread of disinformation, an issue of global concern.
InfiAgent is a new agent framework comparable to GPT4-Agent, developed by replicating Codex. It includes InfiCoder, an open-source model for text-to-code, code-to-code, and freeform code-related QA tasks. The framework focuses on data analysis and integrates an LLM with programming capabilities and a sandbox environment for executing Python code. Why it matters: This research demonstrates the potential for advancements in AI operating systems and highlights areas where current models like GPT-4V can be improved, contributing to the broader development of more capable and versatile AI agents.
A PhD candidate from the University of Waterloo presented on threats from large machine learning systems at MBZUAI. The talk covered data privacy during inference and the misuse of ML systems to generate deepfakes. The speaker also analyzed differential privacy and watermarking as potential solutions. Why it matters: Understanding and mitigating the risks of large ML systems is crucial for responsible AI development and deployment in the region.
Veselin Stoyanov of Tome (previously Facebook AI) gave a talk at MBZUAI on challenges to LLM adoption. The talk covered sparse models, multilingual LLMs, and instruction finetuning. Stoyanov previously led development of RoBERTa, XLM-R, and MultiRay. Why it matters: This talk highlights MBZUAI's role as a forum for discussing key challenges and advancements in large language models, with implications for Arabic NLP and regional AI development.
A CMU professor and MBZUAI affiliated faculty presented research on how LLMs store and use knowledge learned during pre-training. The study used a synthetic biography dataset to show that LLMs may not effectively use memorized knowledge at inference time, even with zero training loss. Data augmentation during pre-training can force the model to store knowledge in specific token embeddings. Why it matters: The research highlights limitations in LLM knowledge manipulation and extraction, with implications for improving model architectures and training strategies for more effective knowledge utilization in Arabic LLMs.
An associate professor of Statistics at the University of Toronto gave a talk on how ensemble learning stabilizes and improves the generalization performance of an individual interpolator. The talk focused on bagged linear interpolators and introduced the multiplier-bootstrap-based bagged least square estimator. The multiplier bootstrap encompasses the classical bootstrap with replacement as a special case, along with a Bernoulli bootstrap variant. Why it matters: While the talk occurred at MBZUAI, the content is about ensemble learning which is a core area for improving AI model performance, and is of general interest to the AI research community.
The InterText project, funded by the European Research Council, aims to advance NLP by developing a framework for modeling fine-grained relationships between texts. This approach enables tracing the origin and evolution of texts and ideas. Iryna Gurevych from the Technical University of Darmstadt presented the intertextual approach to NLP, covering data modeling, representation learning, and practical applications. Why it matters: This research could enable a new generation of AI applications for text work and critical reading, with potential applications in collaborative knowledge construction and document revision assistance.
Daisuke Kihara from Purdue University presented a seminar at MBZUAI on using deep learning for biomolecular structure modeling. His lab is developing 3D structure modeling methods, especially for cryo-electron microscopy (cryo-EM) data. They are also working on RNA structure prediction and peptide docking using deep neural networks inspired by AlphaFold2. Why it matters: Applying advanced deep learning techniques to biomolecular structure prediction can accelerate drug discovery and our understanding of molecular functions.
This article previews a presentation by Kevin Murphy (Google Brain) at MBZUAI on a unified perspective of machine learning, based on his book "Probabilistic Machine Learning: Advanced Topics". The presentation will cover the "4 pillars of ML": predictions, decisions, discovery and generation. Murphy will summarize recent methods and his own contributions in each of these tasks. Why it matters: Hosting prominent international AI researchers strengthens MBZUAI's position as a global hub for AI research and education.
The article provides a basic overview of large language models (LLMs), explaining their functionality and applications. LLMs are AI systems that process and generate human-like text using transformer architecture, trained on vast datasets to predict the next word in a sequence. The piece differentiates between general-purpose, task-specific, and multimodal models, as well as closed-source and open-source LLMs. Why it matters: LLMs are foundational for advancements in Arabic NLP, as evidenced by models like MBZUAI's Jais, and understanding their mechanics is crucial for regional AI development.
A new Bayesian matrix factorization approach is explored for performance prediction in multilingual NLP, aiming to reduce the experimental burden of evaluating various language combinations. The approach outperforms state-of-the-art methods in NLP benchmarks like machine translation and cross-lingual entity linking. It also avoids hyperparameter tuning and provides uncertainty estimates over predictions. Why it matters: Accurate performance prediction methods accelerate multilingual NLP research by reducing computational costs and improving experimental efficiency, especially valuable for Arabic NLP tasks.
Xiaolin Huang from Shanghai Jiao Tong University presented a talk at MBZUAI on training deep neural networks in tiny subspaces. The talk covered the low-dimension hypothesis in neural networks and methods to find subspaces for efficient training. It suggests that training in smaller subspaces can improve training efficiency, generalization, and robustness. Why it matters: Investigating efficient training methods is crucial for resource-constrained environments and can enable broader access to advanced AI.
Thamar Solorio of MBZUAI served as general chair of EMNLP 2024, which hosted over 4,000 attendees. MBZUAI researchers presented nearly 50 studies, including one co-authored by Solorio and Monojit Choudhury that received an Outstanding Paper Award. Key themes included cultural awareness, machine-generated content detection, and LLM empathy and cultural representation. Why it matters: MBZUAI's strong presence at EMNLP highlights its growing influence in the international NLP research community and its focus on culturally aware AI.
MBZUAI researchers presented new resources at EMNLP for improving the factuality of LLMs, including a web application for fact-checking LLM-generated text and benchmarks for evaluating automated fact-checkers. They found that current automated fact-checkers miss nearly 40% of false claims generated by LLMs. The study breaks down the fact-checking process into eight tasks, including decomposition and decontextualization, to identify where systems fail. Why it matters: This work addresses a critical challenge in the deployment of LLMs by providing tools and methods for improving their reliability and trustworthiness, which is essential for widespread adoption in sensitive applications.
Researchers from MBZUAI, University of Washington, and other institutions presented studies at EMNLP 2024 exploring how LLMs represent cultures. A survey analyzed dozens of recent studies on LLMs and culture and proposes a new framework for future research. The survey found that there is no widely accepted definition of 'culture' in NLP, making it challenging to interpret how models represent culture through language. Why it matters: This highlights a key gap in the field and emphasizes the need for a more rigorous and consistent understanding of culture in AI, especially as LLMs become more globally integrated.
Researchers from MBZUAI and Monash University presented a study at EMNLP 2024 examining LLMs' ability to interpret empathy, emotion, and morality in written stories. The study builds on a framework for modeling empathic similarity between narratives, using the EmpathicStories dataset. They are exploring ways to improve LLMs' capabilities with complex concepts like empathy, especially for applications in fields like healthcare. Why it matters: Enhancing LLMs with empathic understanding could lead to more effective and human-centered AI applications, particularly in sensitive domains requiring nuanced communication.
MBZUAI Professor Timothy Baldwin reflects on his decision to move from the University of Melbourne to help build MBZUAI. He emphasizes the importance of inclusive AI education to ensure the benefits of AI are shared globally. Baldwin notes the concentration of AI innovation in a few countries, leading to disparities in language model performance for non-dominant languages. Why it matters: The article highlights MBZUAI's role in addressing the global imbalance in AI development and promoting inclusivity in AI education and research, particularly for Arabic and other underrepresented languages.
MBZUAI researchers presented a study at NAACL 2024 analyzing errors made by open-source LLMs when solving math word problems. The study, led by Ekaterina Kochmar and KV Aditya Srivatsa, investigates characteristics that make math word problems difficult for machines. Llama2-70B was used to test the ability of LLMs to solve these problems, revealing that LLMs can perform math operations correctly but still give the wrong answer. Why it matters: The research aims to improve AI's ability to understand and solve math word problems, potentially leading to better educational applications and teaching methods.
MBZUAI researchers have expanded LLM safety research to Chinese, presenting their work at the 62nd Annual Meeting of the Association for Computational Linguistics in Bangkok. They developed an open-source Chinese dataset of 3,000 prompts translated and localized from the English "Do-Not-Answer" dataset. The dataset includes a "region-specific sensitivity" category to address unique safety risks for Chinese speakers, evaluating if models are over-sensitive in identifying innocuous questions as harmful. Why it matters: This research addresses a critical gap in LLM safety evaluation, ensuring that language models are both safe and effective for diverse linguistic and cultural contexts, particularly in regions with unique sensitivities.
MBZUAI researchers presented a new uncertainty quantification method at ACL to identify hallucinations in LLMs, called claim conditioned probability (CCP). CCP leverages the internal token probabilities generated by the LLM itself to highlight claims with low confidence. Unlike external fact-checking methods, CCP is computationally efficient as it uses probabilities already computed by the model. Why it matters: This research offers a practical approach to mitigate the impact of LLM hallucinations by highlighting potentially unreliable information, improving the trustworthiness of these models, especially for Arabic LLMs.