1 Million Reward for the Comedy King of AI! Beijing University of Posts and Telecommunications, Nanyang Technological University, and others release the 'FunQA' dataset of 'silly videos': Teaching algorithms to learn human humor

Meta AI Releases New Video Learning Model V-JEPA: A Breakthrough in Video Understanding

Recently, the Meta AI team launched the video joint embedding prediction architecture (V-JEPA) model, an innovative initiative aimed at advancing machine intelligence. Humans can naturally process information from visual signals and recognize surrounding objects and motion patterns. An important goal of machine learning is to reveal the fundamental principles that drive unsupervised learning in humans. Researchers proposed a key hypothesis—the predictive feature principle—arguing that the representations of continuous sensory inputs should be able to predict each other. Early research methods utilized slow feature analysis.

Chinese Research Team Unveils VideoChat-Flash, Boosting Long Video Processing Speed by 100 Times

Traditional video understanding models face numerous challenges when processing long videos, including the complexities of understanding the extended context. Although considerable research has been conducted to enhance video understanding capabilities, effectively overcoming the issues of low training and inference efficiency remains difficult. To address these challenges, the research team utilized HiCo technology to compress redundant parts of video information, significantly reducing computational demands while retaining key information. Specifically, HiCo achieves hierarchical compression of the video by segmenting long videos into shorter clips, thereby reducing processing time.

Integrated AI Framework Sa2VA: Achieving Deep Understanding of Images and Videos

Driven by multimodal large language models (MLLMs), significant advancements have been made in tasks related to images and videos, including visual question answering, narrative generation, and interactive editing. However, achieving fine-grained understanding of video content still poses major challenges. These challenges involve tasks such as pixel-level segmentation, tracking with language descriptions, and visual question answering based on specific video prompts. Although current state-of-the-art video perception models excel in segmentation and tracking tasks, they still fall short in open language understanding and conversational capabilities.

Twelve Labs Launches Multimodal Video Understanding AI to Address Video Content Search and Analysis Challenges

NVIDIA Launches Major Breakthrough: AI Video Understanding that Enables Machines to Truly Comprehend Video Content

A Dark Horse in Video Understanding: The Video-XL Model Can Handle Videos Up to One Hour Long!

ZhiYuan Launches Hour-Level Ultra-Long Video Understanding Model Video-XL

The Beijing ZhiYuan Artificial Intelligence Research Institute, in collaboration with Shanghai Jiao Tong University, Renmin University of China, Peking University, and Beijing University of Posts and Telecommunications, has launched an ultra-long video understanding model named Video-XL. This model is an important demonstration of the core capabilities of multimodal large models and a key step towards General Artificial Intelligence (AGI). Compared to existing multimodal large models, Video-XL shows superior performance and efficiency in processing long videos exceeding 10 minutes.

Salesforce AI Research Unveils New Multimodal Model BLIP-3-Video: Cost-Effective Video Understanding

Zhiyuan Releases Native Multimodal World Model Emu3: Achieving Text, Image, and Video Understanding and Generation Solely Through Next Token Prediction

Shusheng · Puyu Lingbi Multimodal Model Upgrade Version 2.5 Supports Longer Contexts and Image-Video Understanding Comparable to GPT-4V

Shusheng · Puyu Lingbi (InternLM-XComposer) Version 2.5 was developed by the Shanghai Artificial Intelligence Laboratory, focusing on long context input and output capabilities, operating smoothly within a length of 96K, and trained with 24K interleaved image-text data. Key upgrades include: high-resolution image understanding, fine-grained video understanding, and multi-turn multi-image dialogue. In application, it can create web pages and write high-quality text-image articles. Evaluations show it surpasses state-of-the-art open-source models across 16 benchmark tests and performs at par with key tasks compared to GPT-4V and Gem.