Search for: All records

Sign language is a priceless means of communication for deaf and hard-of-hearing people to fully enable them to participate in society and interact with others. This study introduces a novel universal sign language system that uses the Gesture-script to generate a detailed description of gestures in videos, which involve continuous movement of hands, arms, heads, and body language. Subsequently, we input this description into a Large Language Model (LLM) to interpret sign language. We deployed a few-shot prompting technique for LLM, enabling it to precisely transfer the sign videos into corresponding sentences in natural language. Furthermore, the Few-shot prompting technique enables our system to interpret multiple types of sign language without pre-training or fine-tuning. 

Abstract The ability to harness the optical or electrical properties of nanoscale particles depends on their assembly in terms of size and spatial characteristics which remains challenging due to lack of size focusing. Electrons provide a clean and focusing agent to initiate the assembly of nanoclusters or nanoparticles. Here an intriguing route is demonstrated to lace gold nanoclusters and nanoparticles in string assembly through electron‐initiated nucleation and aggregative growth of Au(I)‐thiolate motifs on a thin film substrate. This size‐focused assembly is demonstrated by controlling the electron dose under transmission electron microscopic imaging conditions. The Au(I)‐thiolate motifs, in combination with the molecularly mediated alignment, facilitate the interstring electrostatic and intrastring aurophilic interactions, which functions as a molecular template to aid electron‐initiated 1D lacing. The findings demonstrate a hierarchical route for the 1D assemblies with size and spatial tunable catalytic, optical, sensing, and diagnostic properties.