About
Research & Development
Ming-UniAudio: A Unified AI Model for Comprehensive Speech Tasks
TLDR: Ming-UniAudio is a new speech AI framework that unifies understanding, generation, and free-form editing of speech using a novel continuous tokenizer called MingTok-Audio....
Research & Development
Precision Screening for Diabetic Retinopathy Using Deep Ensembles
TLDR: A new AI framework uses an ensemble of seven deep learning models combined with an accuracy-weighted voting system and an entropy-guided uncertainty measure...
Research & Development
Lookahead Unmasking: A New Strategy for Accurate Text Generation in Diffusion Language Models
TLDR: Lookahead Unmasking (LookUM) is a novel, unsupervised inference-time framework for Masked Diffusion Models (MDMs) that improves text generation accuracy by reformulating unmasking as...
Research & Development
Adapting Vision-Language Models for Cell Detection in Optical Microscopy
TLDR: This research introduces Micro-OD, a new benchmark of 252 microscopy images across 11 cell types, to evaluate how Vision-Language Models (VLMs) perform few-shot...
Research & Development
CoPRIS: Accelerating Large Language Model Training with Smart Concurrency and Importance Sampling
TLDR: CoPRIS is a new reinforcement learning framework designed to make training large language models (LLMs) faster and more stable. It addresses the inefficiencies...
Research & Development
AI-Powered Patch Offers Real-Time Brain Oxygenation Monitoring for Neurological Emergencies
TLDR: A new AI-enhanced high-density Near-Infrared Spectroscopy (NIRS) system has been developed to provide accurate, real-time, layer-specific brain oxygenation data, particularly for neurological emergencies....