News

Get ready to be amazed! Researchers at ReLER, CCAI, and Zhejiang University have introduced Human101, a groundbreaking artificial intelligence framework that promises to revolutionize the way we create dynamic, high-fidelity digital human representations from limited data sources such as single-view videos. This innovative approach is geared towards the rapid and efficient reconstruction of 3D digital…

We are absolutely delighted to share with you the groundbreaking research of the researchers from Tsinghua University who have introduced 'LLM4VG': a novel AI benchmark for evaluating Large Language Models (LLMs) on video grounding tasks. This benchmark proposes a dual approach to assess the effectiveness of LLMs in accurately pinpointing specific video segments based on…

Welcome to the 2023 season finale of our weekly roundup of AI news! This past year has been nothing short of an AI rollercoaster ride with awe-inspiring tech advances and exciting news. From the early iterations of ChatGPT to the incredible GPT-4 with vision, AI research has come a long way. We’ve been able to…

We are thrilled to announce the groundbreaking research from AWS AI that introduces a powerful machine learning data augmentation pipeline leveraging controllable diffusion models and CLIP for enhanced object detection! This revolutionary technique has the potential to quickly improve performance in downstream tasks with both sparse and complete datasets. By utilizing diffusion-based inpainting techniques to…

Behold! A revolutionary AI paper from researchers at the University of California San Diego (UCSD) and Johns Hopkins University (JHU) has unveiled the LAW framework: a groundbreaking approach to machine reasoning that integrates language, agent, and world models. This interdisciplinary framework draws upon cognitive science, linguistics, computer science, and artificial intelligence to develop more robust…

We are excited to announce the amazing research presented by the Purdue University researchers with their novel approach, Graph-Based Topological Data Analysis (GTDA), to simplify interpreting complex predictive models like deep neural networks. These models often pose challenges in understanding and generalization, but the GTDA utilizes topological data analysis to transform intricate prediction landscapes into…

Behold! A revolutionary AI paper from researchers at the University of California San Diego (UCSD) and Johns Hopkins University (JHU) has unveiled the LAW framework: a groundbreaking approach to machine reasoning that integrates language, agent, and world models. This interdisciplinary framework draws upon cognitive science, linguistics, computer science, and artificial intelligence to develop more robust…

2023 has been an incredibly exciting and prosperous year for the tech industry, with major players such as Microsoft, Google, and Amazon leading the charge in investing, especially in the realm of generative AI startups. This boom in AI technology has been a major factor in the post-Covid recovery of these tech giants and their…

We are thrilled to share the incredible news that AI-assisted colonoscopies can improve the detection of polyps, particularly when performed by less experienced doctors. This groundbreaking research, conducted by the Chinese University of Hong Kong’s Faculty of Medicine, could pave the way for more accurate diagnosis of colorectal cancers. Colonoscopies, which involve the insertion of an…

We are thrilled to share the incredible news of the development of Fairy, an innovative and highly efficient video-to-video synthesis framework created by the Meta GenAI researchers! This groundbreaking AI-enabled system is designed specifically for instruction-guided video editing tasks, revolutionizing the way video editing is done. With Fairy, users are able to generate high-resolution videos…

Behold the power of large language models (LLMs), such as GPT-4, that can generate advanced texts and execute complex tasks! Their integration into diverse applications from customer service to content creation has been widespread, but with it comes pressing concerns about their potential misuse and implications for digital security and ethics. As a result, the…

We are thrilled to share the incredible news of the development of Fairy, an innovative and highly efficient video-to-video synthesis framework created by the Meta GenAI researchers! This groundbreaking AI-enabled system is designed specifically for instruction-guided video editing tasks, revolutionizing the way video editing is done. With Fairy, users are able to generate high-resolution videos…