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:
-
- 'Looking for a needle in a haystack' out! 'Counting stars' becomes a more accurate method for measuring text length, from Goose Factory
- There is a new method for testing long text ability in large models! Tencent MLPD Lab uses a new open source "counting stars" method to replace the traditional "needle in a haystack" test. In contrast, the new method pays more attention to the examination of the model's ability to handle long dependencies, and the evaluation of the model is more comprehensive and accurate. Using this method, the researchers conducted a "star counting" test on GPT-4 and the well-known domestic KimiChat. As a result, under different experimental conditions, the two models had their own advantages and disadvantages, but both demonstrated strong long text capabilities. △The horizontal axis is a logarithmic coordinate with base 2. So, what kind of test is "counting stars"? More accurate than "finding a needle in a haystack" First, the researchers chose a long text as the context, and the length gradually increased during the test.
- AI 662 2024-04-02 11:55:30
-
- Six business intelligence challenges IT teams must address
- Business intelligence (BI) enables businesses to derive insights from large amounts of data. But doing so requires overcoming a number of strategic and tactical challenges. Currently, organizations of all types are inundated with data from a variety of sources, and trying to make sense of it all is overwhelming. Therefore, a strong business intelligence (BI) strategy can help organize processes and ensure business users are able to access and act on business insights. Through BI strategies, various data sources can be integrated to provide users with accurate and useful information. The benefits of a BI strategy are many. First, it helps organizations better understand their business data and provide deep insights. Second, a BI strategy can also help organizations manage and analyze large amounts of data, according to Seattle-based Launch Consulting Group
- AI 585 2024-04-02 11:52:18
-
- RV fusion performance is amazing! RCBEVDet: Radar also has spring, the latest SOTA!
- Written above & the author’s personal understanding is that the main issue this discussion paper focuses on is the application of 3D target detection technology in the process of autonomous driving. Although the development of environmental vision camera technology provides high-resolution semantic information for 3D object detection, this method is limited by issues such as the inability to accurately capture depth information and poor performance in bad weather or low-light conditions. In response to this problem, the discussion proposed a new multi-mode 3D target detection method-RCBEVDet that combines surround-view cameras and economical millimeter-wave radar sensors. This method provides richer semantic information and a solution to problems such as poor performance in bad weather or low-light conditions by comprehensively using information from multiple sensors. To address this issue, the discussion proposed a method that combines surround-view cameras
- AI 460 2024-04-02 11:49:33
-
- Exploring Siamese networks using contrastive loss for image similarity comparison
- Introduction In the field of computer vision, accurately measuring image similarity is a critical task with a wide range of practical applications. From image search engines to facial recognition systems and content-based recommendation systems, the ability to effectively compare and find similar images is important. The Siamese network combined with contrastive loss provides a powerful framework for learning image similarity in a data-driven manner. In this blog post, we will dive into the details of Siamese networks, explore the concept of contrastive loss, and explore how these two components work together to create an effective image similarity model. First, the Siamese network consists of two identical subnetworks that share the same weights and parameters. Each sub-network encodes the input image into a feature vector, which
- AI 1160 2024-04-02 11:37:12
-
- Alibaba 7B multi-modal document understanding large model wins new SOTA
- New SOTA for multimodal document understanding capabilities! Alibaba's mPLUG team released the latest open source work mPLUG-DocOwl1.5, which proposed a series of solutions to address the four major challenges of high-resolution image text recognition, general document structure understanding, instruction following, and introduction of external knowledge. Without further ado, let’s look at the effects first. One-click recognition and conversion of charts with complex structures into Markdown format: Charts of different styles are available: More detailed text recognition and positioning can also be easily handled: Detailed explanations of document understanding can also be given: You know, "Document Understanding" is currently An important scenario for the implementation of large language models. There are many products on the market to assist document reading. Some of them mainly use OCR systems for text recognition and cooperate with LLM for text processing.
- AI 485 2024-04-02 11:31:27
-
- Is Mamba comparable to Transformer effective on time series?
- Mamba is one of the most popular models recently, and is considered by the industry to have the potential to replace Transformer. The article introduced today explores whether the Mamba model is effective in time series forecasting tasks. This article first introduces the basic principles of Mamba, and then combines this article to explore whether Mamba is effective in time series prediction scenarios. The Mamba model is a deep learning-based model that adopts an autoregressive architecture that can capture long-term dependencies in time series data. Compared with traditional models, the Mamba model performs well on time series forecasting tasks. Through experiments and comparative analysis, this paper found that the Mamba model has good results in time series prediction tasks. it can be accurate
- AI 1118 2024-04-02 11:31:19
-
- Point cloud registration is inescapable for 3D vision! Understand all mainstream solutions and challenges in one article
- Point cloud, as a collection of points, is expected to bring about a change in acquiring and generating three-dimensional (3D) surface information of objects through 3D reconstruction, industrial inspection and robot operation. The most challenging but essential process is point cloud registration, i.e. obtaining a spatial transformation that aligns and matches two point clouds obtained in two different coordinates. This review introduces the overview and basic principles of point cloud registration, systematically classifies and compares various methods, and solves the technical problems existing in point cloud registration, trying to provide academic researchers outside the field and Engineers provide guidance and facilitate discussions on a unified vision for point cloud registration. The general method of point cloud acquisition is divided into active and passive methods. The point cloud actively acquired by the sensor is the active method, and the point cloud is reconstructed later.
- AI 500 2024-04-02 11:31:13
-
- Apple researchers say their on-device model ReALM outperforms GPT-4 and can significantly improve Siri intelligence
- According to news from this site on April 2, although Siri can currently try to describe the pictures in the message, the effect is not stable. However, Apple has not given up exploring the field of artificial intelligence. In a recent research paper, Apple's artificial intelligence team described a model that can significantly improve the intelligence of Siri. They believe that this model, called ReALM, outperformed OpenAI's well-known language model GPT-4.0 in tests. . This article introduces what is special about ReALM, which can simultaneously understand the content on the user's screen and the ongoing operations. Discussions are divided into the following three types: Screen Entity: refers to the content currently displayed on the user's screen. Conversation entity: refers to content related to the conversation. For example, a user says "Call
- AI 1066 2024-04-02 09:16:14
-
- The world's first dual-light source solid-state lidar navigation and obstacle avoidance Roborock V20 sets a new standard for navigation and obstacle avoidance
- On March 29, 2024, the Roborock Technology global conference was held in Beijing. Roborock Technology released two new flagship sweeping and mopping robot products, including the Pioneer flagship Roborock self-cleaning sweeping and mopping robot V20 and the top-tech Roborock self-cleaning sweeping and mopping robot G20S. Among them, Roborock V20 is equipped with a new star array navigation system, which once again refreshes the navigation capabilities of sweeping and mopping robots with the world's first dual-light source solid-state lidar; Roborock G20S fully integrates the pinnacle technology, in the three major areas of cleaning power, versatility and intelligence. In all aspects, it has reached the ceiling in the industry, leading the sweeping robot industry to a comprehensive advanced stage. At the same time, there is also the P10S Pro, which has received enthusiastic response from the market shortly after its launch. In addition to a variety of new sweeping and mopping robot products, Quan Gang, President of Roborock Technology, is launching
- AI 397 2024-04-02 08:25:05
-
- The robot dog died for the first time! US police reveal details
- The US police recently announced a case: the Boston Dynamics robot dog was shot and killed for the first time. Official pictures show that the robot dog has multiple gunshot wounds, the metal shell is dented, and the paint is peeling off, making it no longer usable. The police also spoke highly of the robot dog's sacrifice: it was blocking bullets for human law enforcement partners. Boston Dynamics robot dog was shot dead for the first time, Massachusetts police disclosed details of the case in detail. It was an ordinary Wednesday afternoon when a suspect was hiding inside a home with a gun. The police subsequently dispatched a robot dog named Roscoe and two PacBot510 crawler robots to assist in the search. Under the remote control operation of human soldiers, Roscoe conducted a carpet search of the house. As a result, just after the robot dog finished checking the basement,
- AI 611 2024-04-01 20:25:17
-
- Detailed explanation of rotational position encoding RoPE commonly used in large language models: why is it better than absolute or relative position encoding?
- Since the "AttentionIsAllYouNeed" paper published in 2017, the Transformer architecture has been a cornerstone of the natural language processing (NLP) field. Its design has remained largely unchanged for years, with 2022 marking a major development in the field with the introduction of Rotary Position Encoding (RoPE). Rotated position embedding is the state-of-the-art NLP position embedding technique. Most popular large-scale language models (such as Llama, Llama2, PaLM and CodeGen) already use it. In this article, we’ll take a deep dive into what rotational position encodings are and how they neatly combine the benefits of absolute and relative position embeddings. The need for positional encoding in order to understand Ro
- AI 387 2024-04-01 20:19:01
-
- Google is ecstatic: JAX performance surpasses Pytorch and TensorFlow! It may become the fastest choice for GPU inference training
- The performance of JAX, promoted by Google, has surpassed that of Pytorch and TensorFlow in recent benchmark tests, ranking first in 7 indicators. And the test was not done on the TPU with the best JAX performance. Although among developers, Pytorch is still more popular than Tensorflow. But in the future, perhaps more large models will be trained and run based on the JAX platform. Models Recently, the Keras team benchmarked three backends (TensorFlow, JAX, PyTorch) with the native PyTorch implementation and Keras2 with TensorFlow. First, they select a set of mainstream
- AI 1169 2024-04-01 19:46:11
-
- Learn how to improve coding performance based on GenAI in one article
- Hellofolks, my name is Luga, and today we will talk about technologies related to the artificial intelligence (AI) ecological field - GenAI. Facing the challenges of rapid technological innovation and differentiated business scenarios, traditional coding methods have begun to become acclimated and cannot fully cope with the growing demands. At the same time, emerging general-purpose GenAI (artificial intelligence technology) has great potential to meet this demand. As a representative of artificial intelligence technology, GenAI has begun to be widely used in all walks of life with its strong potential and capabilities. It can automatically learn and adapt to coding needs in different scenarios, greatly improving coding efficiency and quality. Through deep learning and model optimization, GenAI is able to accurately understand different
- AI 818 2024-04-01 18:49:14
-
- Adopting generative AI systems could transform enterprise cloud architectures
- From data availability and security to large language models and selection and monitoring, enterprise adoption of generative AI means the need to reexamine their cloud architecture. Therefore, many companies are rebuilding their cloud architecture and developing generative artificial intelligence systems. So, what changes do these enterprises need to make? What are the emerging best practices? Industry experts said that in the past 20 years, especially in the past two years, he has helped enterprises build some such platforms. Here are his Some advice for enterprises: Understand your own use cases Enterprises need to clearly define the purpose and goals of generative AI in cloud architecture. If you see some false feedback, it's because they don't understand what it means to generate artificial intelligence in business systems. Businesses need to understand their goals
- AI 355 2024-04-01 17:34:12
-
- 0 threshold for free commercial use! Mencius 3-13B large model is officially open source and trained with trillions of token data
- Lanzhou Technology officially announced: The Mencius 3-13B large model is officially open source! This large, cost-effective lightweight model is fully open to academic research and supports free commercial use. Mencius 3-13B has shown good performance in various benchmark evaluations such as MMLU, GSM8K, and HUMAN-EVAL. Especially in the field of lightweight large models with parameters within 20B, his Chinese and English language skills are particularly outstanding. Mathematics and programming skills are also at the forefront. △The above results are based on 5-shot. According to reports, the Mencius 3-13B large model is based on the Llama architecture, and the data set size is as high as 3TTokens. The corpus is selected from web pages, encyclopedias, social media, media, news, and high-quality open source data sets. By in trillion toke
- AI 606 2024-04-01 17:01:22