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:
-
- Quark's large model helps improve academic research efficiency, and its four major advantages improve the accuracy of knowledge
- When rigorous academic research and innovative large-scale model technology are combined, what kind of sparks will be created? Recently, the Quark large model has become the "student master" among domestic large models with its excellent performance as soon as it was launched. At the "Large Model Application Scenario Seminar" recently hosted by the China Association for Science and Technology, the efficiency improvement potential of Quark's large model in four major scenarios, including fast reading and creation and polishing, aroused heated discussions and expectations among scientific and technological workers. . Currently, large model technology is in a stage of rapid development. To promote large models to the application level in the academic field, it has become an industry consensus that the problem of knowledge correctness must first be solved. With long-term accumulation in search business and intelligent technology, the recently released Quark model uses data, platform,
- AI 566 2024-08-09 10:33:50
-
- Training with millions of crystal data to solve the crystallographic phase problem, the deep learning method PhAI is published in Science
- Editor |KX To this day, the structural detail and precision determined by crystallography, from simple metals to large membrane proteins, are unmatched by any other method. However, the biggest challenge, the so-called phase problem, remains retrieving phase information from experimentally determined amplitudes. Researchers at the University of Copenhagen in Denmark have developed a deep learning method called PhAI to solve crystal phase problems. A deep learning neural network trained using millions of artificial crystal structures and their corresponding synthetic diffraction data can generate accurate electron density maps. The study shows that this deep learning-based ab initio structural solution method can solve the phase problem at a resolution of only 2 Angstroms, which is equivalent to only 10% to 20% of the data available at atomic resolution, while traditional ab initio Calculation
- AI 807 2024-08-08 21:22:30
-
- ICML 2024 | Character interaction images, now I understand your prompt words better, Peking University launches a character interaction image generation framework based on semantic perception
- 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 first author and corresponding author of this paper are both from the MIPL Laboratory of Wangxuan Institute of Computer Science, Peking University. The first author is doctoral student Xu Zhu, and the corresponding author is doctoral student Mentor Liu Yang. In recent years, MIPL laboratory has participated in IJCV, CVPR, AAAI, I
- AI 795 2024-08-08 21:01:32
-
- Ng Enda teaches in person, LLM serves as 'teaching assistant', Python programming course suitable for beginners is online
- Professor Ng Enda has opened a new class, and he still teaches in person! Today, Andrew Ng, a famous scholar of artificial intelligence and professor at Stanford University, posted on the social platform X to introduce a new course - AIPython for Beginners, which aims to teach Python programming knowledge from scratch and is suitable for beginners to learn. Andrew Ng will serve as the lecturer of this course. Fenye course link: https://www.deeplearning.ai/short-courses/ai-python-for-beginners/AIPythonforBeginners The course consists of four parts, including: AIPython programming basics: learning
- AI 422 2024-08-08 19:53:20
-
- Llama3 training crashes every 3 hours? Big bean bag model and HKU team improve crispy Wanka training
- As the iteration speed of large models becomes faster and faster, the scale of training clusters becomes larger and larger, and high-frequency software and hardware failures have become a pain point that hinders the further improvement of training efficiency. The checkpoint system is responsible for state storage and recovery during the training process. , has become the key to overcoming training failures, ensuring training progress and improving training efficiency. Recently, the ByteDance Beanbao model team and the University of Hong Kong jointly proposed ByteCheckpoint. This is a large model Checkpointing system native to PyTorch, compatible with multiple training frameworks, and supports efficient reading and writing of Checkpoint and automatic re-segmentation. Compared with existing methods, it has significant performance improvements and ease-of-use advantages. This article introduces large model training
- AI 677 2024-08-08 19:51:02
-
- Hongmeng Smart Travel S9 and full-scenario new product launch conference, a number of blockbuster new products were released together
- This afternoon, Hongmeng Zhixing officially welcomed new brands and new cars. On August 6, Huawei held the Hongmeng Smart Xingxing S9 and Huawei full-scenario new product launch conference, bringing the panoramic smart flagship sedan Xiangjie S9, the new M7Pro and Huawei novaFlip, MatePad Pro 12.2 inches, the new MatePad Air, Huawei Bisheng With many new all-scenario smart products including the laser printer X1 series, FreeBuds6i, WATCHFIT3 and smart screen S5Pro, from smart travel, smart office to smart wear, Huawei continues to build a full-scenario smart ecosystem to bring consumers a smart experience of the Internet of Everything. Hongmeng Zhixing: In-depth empowerment to promote the upgrading of the smart car industry Huawei joins hands with Chinese automotive industry partners to provide
- AI 1196 2024-08-08 07:02:32
-
- Li Feifei personally wrote an article, and dozens of scientists signed a joint letter opposing California's AI restriction bill.
- Is AI really dangerous enough to warrant such regulation? In Silicon Valley, a hot spot for innovation, AI scientists such as Li Feifei and Ng Enda are engaged in a tug-of-war with regulatory authorities over safety and innovation. At the center of this tug of war is a bill called SB-1047. The full name of the bill is "Safe and Secure Innovation for Frontier Artificial Intelligence Act ("Frontier Artificial Intelligence Model Security Innovation Act"), which attempts to establish clear safety standards for high-risk AI models to prevent them from being abused or causing catastrophic consequences. The bill was introduced in the Senate in February this year and subsequently caused great controversy. Many scientists believe that law
- AI 976 2024-08-08 04:21:12
-
- Nature sub-journal, Peking University team's general AI framework conducts comprehensive structure prediction for protein-protein docking, bridging the gap between experiment and calculation
- Editor | Structural prediction of radish skin protein complexes plays an important role in drug development, antibody design and other applications. However, due to limited prediction accuracy, prediction results are often inconsistent with experimental results. A research team from Peking University, Changping Laboratory, and Harvard University proposes ColabDock, a general framework that employs deep learning structural prediction models to integrate different forms and sources of experimental constraints without further large-scale retraining or fine-tuning. . ColabDock outperforms HADDOCK and ClusPro, which use AlphaFold2 as a structure prediction model, not only in complex structure predictions with simulated residues and surface constraints, but also in NMR chemical shift perturbations and covalent labeling.
- AI 537 2024-08-08 02:16:01
-
- AI in use | AI created a life vlog of a girl living alone, which received tens of thousands of likes in 3 days
- Editor of the Machine Power Report: Yang Wen The wave of artificial intelligence represented by large models and AIGC has been quietly changing the way we live and work, but most people still don’t know how to use it. Therefore, we have launched the "AI in Use" column to introduce in detail how to use AI through intuitive, interesting and concise artificial intelligence use cases and stimulate everyone's thinking. We also welcome readers to submit innovative, hands-on use cases. Video link: https://mp.weixin.qq.com/s/2hX_i7li3RqdE4u016yGhQ Recently, the life vlog of a girl living alone became popular on Xiaohongshu. An illustration-style animation, coupled with a few healing words, can be easily picked up in just a few days.
- AI 733 2024-08-07 22:53:33
-
- iFlytek releases smart office notebook Air 2 to create an office AI assistant that understands you better
- The AI assistant that understands you better makes working more comfortable. On August 6, iFlytek held a new smart office notebook product launch conference titled "AI Understands You, Makes You Light and Comfortable", and officially released the Air2 series of iFlytek smart office laptops. Air2 deeply integrates iFlytek's Spark large model capabilities, bringing three major AI skills: meeting minutes, note analysis, and AI writing; Air2 also uses the semantic turning point recognition algorithm for the first time, greatly improving the accuracy of role-based transcription for multi-person meetings; in addition , Air2 also has a large visual screen-to-body ratio of 88.5%, which is thin and portable while bringing a "big" experience with the ink screen field of view. Thin, light and ultra-portable, 88.5% large screen-to-body ratio Air2 uses an 8.2-inch EInk ink screen, super large
- AI 1026 2024-08-07 22:46:03
-
- Alibaba International will launch the world's first AI-driven B2B search engine
- Following the recent release of SearchGPT by OpenAI, the ignited "AISearch" track welcomes a new heavyweight player: the AI-driven B2B search engine developed by Alibaba International will be officially unveiled in September this year. At present, the emergence of AI has completely changed the search experience. Alibaba International's new product will be the world's first professional version of the AI search engine. Its application scenarios and business value are very clear: it focuses on global B2B e-commerce and is aimed at small and medium-sized enterprise owners around the world. It is intended to change the global procurement process and enable It is more intuitive and efficient. It can actively understand the natural language of buyers and convert it into professional purchasing requests. Furthermore, it can also predict demand and provide suggestions based on global market data to achieve better results.
- AI 1150 2024-08-07 20:54:30
-
- ACM MM2024 | NetEase Fuxi's multimodal research gained international recognition again, promoting new breakthroughs in cross-modal understanding in specific fields
- 1. The 32nd ACM International Conference on Multimedia (ACM MM) announced the acceptance results of papers. NetEase Fuxi’s latest research result "Selection and Reconstruction of Key Locals: A Novel Specific Domain Image-Text Retrieval Method" was selected. . The research directions of this paper involve visual language pre-training (VLP), cross-modal image and text retrieval (CMITR) and other fields. This selection marks the multi-modal capabilities of NetEase Fuxi Lab
- AI 1232 2024-08-07 20:16:12
-
- If you think Python is slow, this AI analyzer will help you detect and suggest optimizations, and won the best paper in OSDI
- In recent years, there is a programming language that is ubiquitous and everyone knows its name. It is even preached on the Internet that one can become a programming master by learning it. It is Python. Learning Python is like a treasure: it is easy to read and write, has strong compatibility, rich libraries, and a wide range of uses. It has become very popular in the era of data and machine learning. But as good as it is, Python still has its own shortcomings, that is, it is too slow! There are many programming languages, such as C++, Fortran and Java, which are more famous among them. Compared with them, programs written in Python are notoriously slow. University ofMassachusettsAmherst
- AI 1084 2024-08-07 19:39:13
-
- PRO | Why are large models based on MoE more worthy of attention?
- In 2023, almost every field of AI is evolving at an unprecedented speed. At the same time, AI is constantly pushing the technological boundaries of key tracks such as embodied intelligence and autonomous driving. Under the multi-modal trend, will the situation of Transformer as the mainstream architecture of AI large models be shaken? Why has exploring large models based on MoE (Mixed of Experts) architecture become a new trend in the industry? Can Large Vision Models (LVM) become a new breakthrough in general vision? ...From the 2023 PRO member newsletter of this site released in the past six months, we have selected 10 special interpretations that provide in-depth analysis of technological trends and industrial changes in the above fields to help you achieve your goals in the new year. be prepared. This interpretation comes from Week50 2023
- AI 714 2024-08-07 19:08:10
-
- CVPR 2024 | Four-dimensional space-time pre-training of autonomous driving world model
- Peking University and the EVLO innovation team jointly proposed DriveWorld, a four-dimensional space-time pre-training algorithm for autonomous driving. This method uses a world model for pre-training, designs a memory state space model for four-dimensional spatio-temporal modeling, and reduces the random uncertainty and knowledge uncertainty faced by autonomous driving by predicting the occupation grid of the scene. This paper has been accepted by CVPR2024. Paper title: DriveWorld: 4DPre-trainedSceneUnderstandingviaWorldModelsforAutonomousDriving Paper link: https://arxiv.org/abs/2405.04390 1. Motion
- AI 1014 2024-08-07 19:01:41