current location:Home > Technical Articles > Technology peripherals > AI
- Direction:
- All web3.0 Backend Development Web Front-end Database Operation and Maintenance Development Tools PHP Framework Daily Programming WeChat Applet Common Problem Other Tech CMS Tutorial Java System Tutorial Computer Tutorials Hardware Tutorial Mobile Tutorial Software Tutorial Mobile Game Tutorial
- Classify:
-
- The multimodal model evaluation framework lmms-eval is released! Comprehensive coverage, low cost, zero pollution
- The AIxiv column is a column where this site publishes academic and technical content. In the past few years, the AIxiv column of this site has received more than 2,000 reports, covering top laboratories from major universities and companies around the world, effectively promoting academic exchanges and dissemination. If you have excellent work that you want to share, please feel free to contribute or contact us for reporting. Submission email: liyazhou@jiqizhixin.com; zhaoyunfeng@jiqizhixin.com With the deepening of research on large models, how to promote them to more modalities has become a hot topic in academia and industry. Recently released closed-source large models such as GPT-4o, Claude3.5, etc. already have super image understanding capabilities, LLaV
- AI 536 2024-08-21 16:38:07
-
- NVIDIA's first AI NPC enters the game, a domestic masterpiece, the 4B model only requires 2G of video memory
- Players are all asking: When will the game be launched? The large model-driven game NPC has finally been launched. Early this morning, Nvidia released a game demo. Now when playing games, you can use voice dialogue to communicate with NPCs, understand the level objectives, optimize equipment configuration, and then adjust the weapon color to start the battle. The highly anticipated AINPC engine has made its debut in the multiplayer mecha combat game MechaBREAK. The first time in the world, it is a domestic game. As a PVP-heavy game, Unlimited Machine aims to become a global e-sports product. It just ended its pre-test in August, and many people said that "the closed beta ended too early." At present, the game has more than one million wishlist collections on the Steam platform, and many game videos have appeared at home and abroad.
- AI 487 2024-08-21 16:37:00
-
- Supporting 1024 frames and nearly 100% accuracy, NVIDIA 'LongVILA” begins to develop long videos
- Now, Long-Context Visual Language Model (VLM) has a new full-stack solution - LongVILA, which integrates system, model training and data set development. At this stage, it is very important to combine the multi-modal understanding of the model with the long-context capability. The basic model that supports more modalities can accept more flexible input signals so that people can interact with the model in more diverse ways. . And longer context allows the model to process more information, such as long documents and long videos. This ability also provides the functionality needed for more real-world applications. However, the problem currently faced is that some work has enabled long-context visual language models (VLM), but usually in a simplified approach rather than providing a full
- AI 662 2024-08-21 16:35:04
-
- It can save 2 million US dollars every year. Without Elsevier, MIT is doing pretty well.
- After negotiations with Elsevier broke down in 2020, MIT took a new route that was not tied to large journal subscriptions, saving $2 million per year. For universities, journal subscription fees have always been a huge expense. Especially the cost of subscribing to journals from the world’s top publishing groups like Elsevier. As we all know, Elsevier is one of the largest publishers of medical and other scientific literature in the world. The well-known academic journals "The Lancet" and "Cell" we are familiar with are all published by Elsevier. However, while providing convenience to authors and readers, Elsevier's subscription fees have always been high, inevitably placing a heavy financial burden on universities and other scientific research institutions. In addition, Elsevier’s publishing nature of mainly closed journals is not in line with scientific research.
- AI 1137 2024-08-21 16:32:11
-
- Joining forces with Science to target young scholars, Chen Tianqiao launches the International Award for AI-Driven Scientific Research
- Recently, the Tianqiao & Chrissy Chen Institute (referred to as the institute) and Science magazine jointly announced the launch of an AI-driven science award to honor young people around the world who use artificial intelligence technology to promote scientific research in various fields and make breakthrough progress. the scientist. Chen Tianqiao, a former Chinese Internet tycoon, has been deeply involved in the field of brain science for 8 years. He has devoted himself to the research and development of new generation artificial intelligence technology for nearly two years and has achieved a lot. The launch of this international award on this occasion is quite eye-catching. Focus on globalization, interdisciplinary, young scientists. According to the official website of the institute and Science magazine, the content of this award collection is about 1,000 words, and the target is the global use of artificial intelligence-related technologies.
- AI 683 2024-08-21 16:31:55
-
- The world's first and exclusive to Denza! BYD's new black technology Yi Sanfang is officially released, reshaping the benchmark for new energy luxury technology
- On August 20, e³|Yisanfang Intelligent Technology was fully launched - the 2024 DENZA Automotive Technology Day and the DENZA Z9GT pre-sale conference were successfully held. DENZA's new exclusive technology "Yisanfang" was shockingly released, and DENZA Z9GT was launched for the first time in the world. . As the world's first vehicle intelligent control technology platform, Yisanfang redefines the new energy luxury technology benchmark with its new intelligent and electronic integration technology. Denza Z9GT and Denza Z9 dual flagships have officially started pre-sale, with the pre-sale price range ranging from 339,800 yuan to 419,800 yuan. They take into account ultimate performance and luxurious luxury, setting a new benchmark for new energy GT with the strength of the pioneer, leading the way Luxury GT enters the new energy era! Tailored for Denza, BYD is another new breakthrough in vehicle intelligence. Vehicle intelligence is truly intelligent. As a luxury brand in the new energy era,
- AI 719 2024-08-21 13:46:00
-
- PICO launches MR mixed reality all-in-one machine PICO 4 Ultra to provide users with a new experience
- On August 20, 2024, PICO held a new product communication meeting in Beijing and officially launched its first MR mixed reality all-in-one machine, PICO4Ultra. The launch of PICO4Ultra marks PICO's new breakthrough in the field of MR technology, bringing users a richer and more personalized mixed reality experience. At the hardware level, PICO4Ultra is equipped with the new Qualcomm Snapdragon® At the system level, PICO4Ultra introduces a panoramic screen workbench, allowing users to open and arrange multiple
- AI 831 2024-08-21 13:43:20
-
- Registration for ByteDance's 2024 scholarship program starts! 100,000 per person to help scientific research for the future!
- Click this link to download the scholarship application form (it is recommended to open it on PC)
- AI 1145 2024-08-20 16:54:44
-
- Mi Kuai AI: The leading startup company in the AI animation track recruits AI algorithm interns/engineers
- 01 Company Vision Miguo AI (miguo.ai) is a Shanghai start-up company with great potential. The founding team comes from Bilibili and Shanghai Jiao Tong University. It combines industry, academia and research, and has a clear business model. It is committed to solving comics and animation problems through AI technology. The company's goal is to become the world's largest AI animation platform by reducing the production capacity and output pain points, reducing the creation threshold and production costs, and improving the creator ecology. 02 Founding team founder Ding Li, former vice president of Bilibili, former CEO of Bilibili Comics, and co-founder of Huya Live. Co-founder Chen Dazhi has 12 years of experience in the field of venture capital, focusing on investment and business research in the cultural and technology industries. Currently responsible for the company's strategic planning, financing and cooperative ecological construction. Technical partner Niu Li (ustcnewl
- AI 358 2024-08-20 16:44:59
-
- Tesla hires 'motion capture artist' to train humanoid robot Optimus, earning up to $48 an hour
- The data for embodied intelligence comes from here. Tomorrow, the annual World Robot Conference will kick off again! Recently, many domestic robot companies have released new products, such as the second-generation humanoid robot from Zhihuijun’s “Zhiyuan Robot” company and the new-generation AI robot assistant Astribot S1 released by Stardust Intelligent. Abroad, Tesla also has new news on the training of its humanoid robot Optimus. The company is recruiting people to don motion-capture suits to imitate intended movements to get the massive amounts of data it needs to train the robots. Tesla defines such people on its job site as "data collection operators," whose primary goals are to collect data, assist with engineering requests and report equipment feedback. It is understood that the hourly salary for this position can be up to 48
- AI 686 2024-08-20 16:42:26
-
- Zhejiang University Li Xi's team: A new method for referring to expression understanding, ScanFormer iterates from coarse to fine to eliminate visual redundancy
- The AIxiv column is a column where this site publishes academic and technical content. In the past few years, the AIxiv column of this site has received more than 2,000 reports, covering top laboratories from major universities and companies around the world, effectively promoting academic exchanges and dissemination. If you have excellent work that you want to share, please feel free to contribute or contact us for reporting. Submission email: liyazhou@jiqizhixin.com; zhaoyunfeng@jiqizhixin.com The authors of this paper are all from the team of Professor Li Xi of Zhejiang University. The first author of the paper is doctoral student Su Wei, and the corresponding author is Professor Li Xi (IETFellow, National Distinguished Young Scholar) ). Professor Li Xi’s team has published in authoritative international journals (such as TPAMI, IJC
- AI 382 2024-08-20 16:35:49
-
- AI designs protein 'switches' from scratch, an amazing breakthrough in protein design, David Baker's research is published in Nature
- Editor | KX In life, it's easy to turn on a lamp or adjust the lighting. But systems that achieve similar control of biomolecular functions are complex and poorly understood. In biology, protein functions are turned on and off in complex ways, and allosteric regulation is one of the important biological regulatory mechanisms that is essential for healthy metabolism and cell signaling. But creating allostery in synthetic protein systems has always presented significant challenges. Recently, David Baker's team at the University of Washington designed a protein that can switch between assembly and disassembly reliably and accurately through allosteric control. The researchers used AI to design new proteins that do not exist in nature and designed a variety of dynamic protein arrangements. David Baker said:
- AI 735 2024-08-20 13:41:30
-
- 'Reconstructing the Future, Industrial Intelligence Leap' 2023 Large Model Industry Frontier Forum is about to open!
- 2023 is the turning point of the artificial intelligence technology revolution and an important engine for a new round of technological innovation and industrial upgrading. Standing at the beginning of the intelligent era opened by the AI large model, a new round of scientific and technological revolution and industrial transformation are evolving in depth. The digital business era driven by digital productivity is bursting with unprecedented development opportunities. On November 23, the 2023 Large Model Industry Frontier Forum, co-sponsored by Zhongguancun Science and Technology, the Artificial Intelligence Innovation Center of China Academy of Information and Communications Technology, and the Key Laboratory of Artificial Intelligence Key Technology and Application Evaluation of the Ministry of Industry and Information Technology, will be held at Park Plaza Hotel in Beijing. Official opening. The theme of this forum is "Reconstructing the Future, Industrial Intelligent Transformation". Representatives from all walks of life from industry, academia, research and investment will focus on "large model industry upgrading".
- AI 691 2024-08-19 18:33:13
-
- Why didn't I know when I was learning line generation: There is an equivalence relationship between matrices and graphs?
- The matrix is difficult to understand, but it may be different if you look at it from another perspective. When learning mathematics, we are often frustrated by the difficulty and abstractness of the knowledge we learn; but sometimes, just by changing the perspective, we can find a simple and intuitive solution to the problem. For example, when we were learning the formula for the sum of squares (a+b)² when we were children, we may not understand why it is equal to a²+2ab+b². We only knew that it was written like this in the book and the teacher asked us to remember it like this; until one day we saw I saw this animated picture: It suddenly dawned on me that we can understand it from a geometric perspective! Now, this sense of enlightenment occurs again: a non-negative matrix can be equivalently converted into the corresponding directed graph! As shown in the figure below, the 3×3 matrix on the left can actually be
- AI 562 2024-08-19 16:52:48
-
- The first large UI model in China is released! Motiff's large model creates the best assistant for designers and optimizes UI design workflow
- Artificial intelligence is developing faster than you might imagine. Since GPT-4 introduced multimodal technology into the public eye, multimodal large models have entered a stage of rapid development, gradually shifting from pure model research and development to exploration and application in vertical fields, and are deeply integrated with all walks of life. In the field of interface interaction, international technology giants such as Google and Apple have invested in the research and development of large multi-modal UI models, which is regarded as the only way forward for the mobile phone AI revolution. In this context, the first large-scale UI model in China was born. On August 17, at the IXDC2024 International Experience Design Conference, Motiff, a design tool in the AI era, launched its independently developed UI multi-modal model - Motiff Model. This is the world's first UI design tool
- AI 927 2024-08-19 16:48:06