News

Natural language processing faces the challenge of precision in language models, with a particular focus on large language models (LLMs). These LLMs often produce factual errors or 'hallucinations' due to their reliance on internal knowledge bases. Retrieval-augmented generation (RAG) was introduced to improve the generation process of LLMs by including external, relevant knowledge. However, RAG’s effectiveness…

3D medical image segmentation faces difficulties in capturing global data from high-resolution images often resulting in suboptimal segmentation. A possible solution involves the use of depth-wise convolution with larger kernel sizes to detect a wider array of features. However, this approach may not fully capture the relations across distant pixels, hence needing a complementary method. Transformers…

The world of artificial intelligence (AI) has seen an impressive paradigm shift with the transition from one foundational model to another. Various models, such as Mamba, Mamba MOE, MambaByte, and more recent methods like Cascade, Layer-Selective Rank Reduction (LASER), and Additive Quantization for Language Models (AQLM), have showcased increased cognitive capabilities. This progression is humorously…

Scientists from Tel-Aviv University and The Open University in Israel have developed DiffMoog, the first comprehensive differentiable modular synthesizer. Designed for automating sound matching and replicating audio input, the synthesizer enhances the capabilities of machine learning and neural networks in sound synthesis. The innovative DiffMoog presents an array of features commonly found in commercial synthesizers,…

The modern digital age sees an escalating demand for smart, effective digital assistants for tasks as varied as communication, learning, research, and entertainment. However, finding digital assistants proficient in multiple languages remains a challenge. In our increasingly globalized world, bilingual or multilingual capabilities are of paramount importance. There are numerous solutions offered by various large language…

Models such as CLIP (Radford et al., 2021) that fuse visual and language data to understand complex tasks show potential but struggle with performance issues when presented with untrained or out-of-distribution (OOD) data. This concern is of particular importance when models encounter novel categories not in their training set, which can pose potential safety issues.…

Human-robot interaction presents numerous challenges, including that of equipping robots with human-like expressive behavior. Traditional rule-based methods require scalability in new social contexts, while data-driven approaches are limited by the need for specific, wide-ranging datasets. As the diversity of social interactions increases, the need for more flexible, context-sensitive solutions intensifies. Generating socially acceptable behaviors for robots…

CDAO Financial Services 2024, to be held on February 13th and 14th in New York, marks its tenth anniversary of bringing together key figures from the financial services data and analytics field. The event anticipates accelerating the growth of data-driven innovation within the industry which artificial intelligence (AI) is progressively shaping. The congregation is set to…

In the advanced AI realm, a considerable obstacle is the data's security and privacy, particularly when using external services. Numerous businesses and individuals have stringent regulations regarding where their sensitive data should be stored and processed. Traditional solutions often necessitate sending data to external servers, sparking worries about compliance with data protection laws and control…

Large language models (LLMs) like GPT-4 require considerable computational power and memory, making their efficient deployment challenging. Techniques like sparsification have been developed to reduce these demands, but can introduce additional complexities like complicated system architecture and partially realized speedup due to limitations in current hardware architectures. Compression methods for LLMs such as sparsification, low-rank approximation,…

Recent advances in machine learning (ML) and artificial intelligence (AI) are being applied across numerous fields thanks to increased computing power, extensive data access, and improved ML techniques. Researchers from MIT and Harvard University have used these advancements to shed light on the brain's reaction to language, using AI models to trigger and suppress responses…

In the advanced AI realm, a considerable obstacle is the data's security and privacy, particularly when using external services. Numerous businesses and individuals have stringent regulations regarding where their sensitive data should be stored and processed. Traditional solutions often necessitate sending data to external servers, sparking worries about compliance with data protection laws and control…