Giovanni Puccetti from ISTI-CNR presented research on linguistic probing of language models like BERT and RoBERTa. The research investigates the ability of these models to encode linguistic properties, linking this ability to outlier parameters. Preliminary work on fine-tuning LLMs in Italian and detecting synthetic news generation was also presented. Why it matters: Understanding the inner workings and linguistic capabilities of LLMs is crucial for improving their reliability and adapting them to diverse languages like Arabic.
Pascal Fua from EPFL presented an approach to implementing convolutional neural nets that output complex 3D surface meshes. The method overcomes limitations in converting implicit representations to explicit surface representations. Applications include single view reconstruction, physically-driven shape optimization, and bio-medical image segmentation. Why it matters: This research advances geometric deep learning by enabling end-to-end trainable models for 3D surface mesh generation, with potential impact on various applications in computer vision and biomedical imaging in the region.
TII's Falcon 40B, a 40-billion-parameter open-source AI model, has ranked #1 on Hugging Face's Open LLM Leaderboard, surpassing models like LLaMA and StableLM. The leaderboard uses benchmarks like AI2 Reasoning Challenge, HellaSwag, MMLU, and TruthfulQA. Trained on one trillion tokens, Falcon 40B's weights are available for research and commercial use. Why it matters: This achievement positions the UAE as a leader in generative AI and promotes transparent, inclusive AI development.
LUMA AI is expanding its presence in Saudi Arabia, establishing its regional headquarters in the Kingdom. The company is partnering with HUMAIN, a Saudi entity, to support the creative industry through AI tools. LUMA AI's technology enables the creation of 3D models from images and videos, catering to the growing demand for digital content in the region. Why it matters: This move signals increasing investment and interest in AI-driven solutions for creative applications within the Saudi Arabian market.
MBZUAI researchers introduce SocialMaze, a new benchmark for evaluating social reasoning capabilities in large language models (LLMs). SocialMaze includes six diverse tasks across social reasoning games, daily-life interactions, and digital community platforms, emphasizing deep reasoning, dynamic interaction, and information uncertainty. Experiments show that LLMs vary in handling dynamic interactions, degrade under uncertainty, but can be improved via fine-tuning on curated reasoning examples.
MBZUAI researchers demonstrated a low-latency, multilingual multimodal AI system at GITEX that integrates speech, text, and visual capabilities for more lifelike human-machine conversation. The demo, led by Dr. Hisham Cholakkal, includes a mobile app where users can point their camera at an object and ask questions, receiving spoken answers in multiple languages. They are also integrating the model into a robot dog that can respond to voice commands. Why it matters: This work addresses key challenges in deploying LLMs to real-world applications in the Middle East, such as multilingual support and real-time responsiveness.