Researchers at King Abdullah University of Science and Technology (KAUST) have developed a wearable microneedle patch capable of continuously measuring drug levels beneath the skin. The device wirelessly transmits real-time data to a smartphone, using tiny microneedles to access interstitial fluid and electrochemical biosensors. The technology was successfully demonstrated by tracking vancomycin concentrations in preclinical studies. Why it matters: This innovation could enable more precise and personalized medicine dosing by providing continuous insight into drug pharmacokinetics, improving safety and efficacy compared to current periodic blood tests.
Technology Innovation Institute (TII) has released Falcon Mamba 7B, a new large language model and the first State Space Language Model (SSLM) in its Falcon series. Falcon Mamba 7B is the top-ranked open-source SSLM globally, outperforming Meta's Llama 3.1 8B, Llama 3 8B, and Mistral’s 7B on HuggingFace benchmarks. SSLMs excel at understanding complex, evolving situations and have applications in NLP tasks like machine translation and text summarization. Why it matters: This release strengthens the UAE's position as an AI hub, demonstrating TII's commitment to pioneering research and open-source AI development in the region.
The researchers introduce KAU-CSSL, the first continuous Saudi Sign Language (SSL) dataset focusing on complete sentences. They propose a transformer-based model using ResNet-18 for spatial feature extraction and a Transformer Encoder with Bidirectional LSTM for temporal dependencies. The model achieved 99.02% accuracy in signer-dependent mode and 77.71% in signer-independent mode, advancing communication tools for the SSL community.
The paper introduces UAE-3D, a multi-modal VAE for 3D molecule generation that compresses molecules into a unified latent space, maintaining near-zero reconstruction error. This approach simplifies latent diffusion modeling by eliminating the need to handle multi-modality and equivariance separately. Experiments on GEOM-Drugs and QM9 datasets show UAE-3D establishes new benchmarks in de novo and conditional 3D molecule generation, with significant improvements in efficiency and quality.
The paper introduces Sparse-Quantized Representation (SpQR), a new compression format and quantization technique for large language models (LLMs). SpQR identifies outlier weights and stores them in higher precision while compressing the remaining weights to 3-4 bits. The method achieves less than 1% accuracy loss in perplexity for LLaMA and Falcon LLMs and enables a 33B parameter LLM to run on a single 24GB consumer GPU. Why it matters: This enables near-lossless compression of LLMs, making powerful models accessible on resource-constrained devices and accelerating inference without significant accuracy degradation.
MBZUAI researchers had 26 papers accepted at ACL 2023, a top NLP conference. Assistant Professor Alham Fikri Aji co-authored eight papers, including one on crosslingual generalization through multitask finetuning (MTF). Deputy Department Chair Preslav Nakov co-authored a paper on a Bulgarian language understanding benchmark dedicated to the memory of Yale Computer Scientist Dragomir R. Radev. Why it matters: MBZUAI's strong presence at ACL highlights its growing influence in the NLP field and its contributions to multilingual AI research.
KAUST showcased its smart health technologies at the Global Health Exhibition in Riyadh, including AI-driven tools for autism diagnosis, SkinGPT-4 for dermatological image analysis, and nanobodies for detecting dengue and Zika viruses. They also presented a wirelessly connected smart bandage for monitoring wound parameters. KAUST signed agreements with Saudi partners for research collaborations during the event. Why it matters: KAUST's efforts in smart health innovation are poised to strengthen healthcare systems in Saudi Arabia and beyond, addressing critical medical challenges through advanced technologies.
KAUST and international collaborators discovered that the Rub’ Al-Khali desert was once a vast lake and river system that supported grasslands and savannahs. The "Green Arabia" period, marked by heavy rainfall, facilitated the growth of a lake covering 1,100 km² with a depth of 42 meters, which overflowed and carved a 150 km valley. The research, published in Communications Earth & Environment, highlights the impact of climate cycles on landscapes and human societies. Why it matters: Understanding the past climate transformations in the Arabian Peninsula is crucial for predicting the consequences of current climate change and its impact on regional habitability.