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:
-
- Caltech Chinese use AI to subvert mathematical proofs! Speed up 5 times shocked Tao Zhexuan, 80% of mathematical steps are fully automated
- LeanCopilot, this formal mathematics tool that has been praised by many mathematicians such as Terence Tao, has evolved again? Just now, Caltech professor Anima Anandkumar announced that the team released an expanded version of the LeanCopilot paper and updated the code base. Image paper address: https://arxiv.org/pdf/2404.12534.pdf The latest experiments show that this Copilot tool can automate more than 80% of the mathematical proof steps! This record is 2.3 times better than the previous baseline aesop. And, as before, it's open source under the MIT license. In the picture, he is Song Peiyang, a Chinese boy. He is
- AI 886 2024-04-23 15:01:29
-
- The latest from Oxford University! Mickey: 2D image matching in 3D SOTA! (CVPR\'24)
- Project link written in front: https://nianticlabs.github.io/mickey/ Given two pictures, the camera pose between them can be estimated by establishing the correspondence between the pictures. Typically, these correspondences are 2D to 2D, and our estimated poses are scale-indeterminate. Some applications, such as instant augmented reality anytime, anywhere, require pose estimation of scale metrics, so they rely on external depth estimators to recover scale. This paper proposes MicKey, a keypoint matching process capable of predicting metric correspondences in 3D camera space. By learning 3D coordinate matching across images, we are able to infer metric relative
- AI 898 2024-04-23 13:20:21
-
- New test benchmark released, the most powerful open source Llama 3 is embarrassed
- If the test questions are too simple, both top students and poor students can get 90 points, and the gap cannot be widened... With the release of stronger models such as Claude3, Llama3 and even GPT-5 later, the industry is in urgent need of a more difficult and differentiated model Benchmarks. LMSYS, the organization behind the large model arena, launched the next generation benchmark, Arena-Hard, which attracted widespread attention. There is also the latest reference for the strength of the two fine-tuned versions of Llama3 instructions. Compared with MTBench, which had similar scores before, the Arena-Hard discrimination increased from 22.6% to 87.4%, which is stronger and weaker at a glance. Arena-Hard is built using real-time human data from the arena and has a consistency rate of 89.1% with human preferences.
- AI 780 2024-04-23 12:13:10
-
- Exploring the Python artifact: How does the eli5 module interpret the prediction results of the machine learning model?
- In the field of Python programming, sometimes we encounter some complex codes or algorithms, and it is difficult to understand the logic and principles. In order to help us better understand the operating mechanism behind the code, the ELI5 module came into being. The ELI5 module is a Python library that can interpret the predictions of machine learning models. Help us understand how the model makes decisions. Through the ELI5 module, we can use the interpreter learning model to understand the predictions of the model. This module provides a concise way to explain the model's decision for a specific sample. The working principle of the ELI5 module is to help us understand the model by sorting and visualizing the importance of features. In this article, we will explore the use of the eli5 module in different scenarios.
- AI 1406 2024-04-23 09:20:02
-
- Tencent's collaborative SaaS products are fully integrated into the Hunyuan model to achieve intelligent upgrades
- On April 22, Tencent announced that its collaborative SaaS products are fully integrated into Tencent's Hunyuan model. In addition to "three outstanding products" such as Enterprise WeChat, Tencent Conference, and Tencent Documents, Tencent Enjoy, Tencent Electronic Signature, Tencent Questionnaire, Collaborative SaaS products such as Tencent Cloud AI Code Assistant have also been upgraded intelligently. Tencent Hunyuan has highlighted its practicality label since its launch, and one of its core missions is to optimize the user experience of Tencent products. Tencent's Hunyuan large model has now expanded to a trillion-level parameter scale. It is the first in China to adopt a hybrid model of experts (MoE) structure. It is better at handling complex scenarios and multi-task scenarios, and its overall Chinese performance is at the leading level in the industry. Tencent Hunyuan has excellent performance in mathematics, coding, logical reasoning, multi-round dialogue and text generation, while providing
- AI 733 2024-04-23 08:13:27
-
- Efficient single-stage short-term RGB-T single target tracking method based on Transformer
- Introduction As shown in Figure 1, the existing three-stage RGB-T single target tracking network usually uses two independent feature extraction branches, which are respectively responsible for extracting the features of the two modalities. However, mutually independent feature extraction branches will lead to a lack of effective information interaction between the two modalities in the feature extraction stage. Therefore, once the network completes offline training, it can only extract fixed features from each modal image and cannot dynamically adjust according to the actual modal state to extract more targeted dynamic features. This limitation restricts the network's ability to adapt to diverse target bimodal appearances and the dynamic correspondence between modal appearances. As shown in Figure 2, this feature extraction method is not suitable for practical application scenarios of RGB-T single target tracking, especially in complex environments, because it is
- AI 967 2024-04-23 08:01:01
-
- OpenAI or DIY? Uncovering the true cost of self-hosting large language models
- Your service standard has been positioned as "AI-driven" by integrating large-scale language models. Your website homepage proudly showcases the revolutionary impact of your AI-driven services through interactive demos and case studies. This is also the first mark your company has left in the global GenAI field. Your small but loyal user base is enjoying an improved customer experience, and you can see potential for future growth. However, as the month enters its third week, you receive an email from OpenAI that surprises you: just a week ago you were talking to customers, assessing product market fit (PMF), and now, thousands Thousands of users flock to your site (anything can go viral on social media these days) and your AI
- AI 1313 2024-04-22 18:01:02
-
- Beyond BEVFusion! DifFUSER: Diffusion model enters autonomous driving multi-task (BEV segmentation + detection dual SOTA)
- Written above & the author’s personal understanding At present, as autonomous driving technology becomes more mature and the demand for autonomous driving perception tasks increases, the industry and academia very much hope for an ideal perception algorithm model that can simultaneously complete three-dimensional target detection and based on Semantic segmentation task in BEV space. For a vehicle capable of autonomous driving, it is usually equipped with surround-view camera sensors, lidar sensors, and millimeter-wave radar sensors to collect data in different modalities. This makes full use of the complementary advantages between different modal data, making the data complementary advantages between different modalities. For example, 3D point cloud data can provide information for 3D target detection tasks, while color image data can provide more information for semantic segmentation tasks. accurate information. Needle
- AI 774 2024-04-22 17:49:07
-
- Uncovering the power of large language models (LLMs): How startups are revolutionizing the way they operate through streamlined integration
- Large language models (LLMs) have become a game-changer for businesses of all sizes, but their impact on startups has been especially dramatic. To understand why, let’s take a look at what advantages startups have over established players and why AI is an important enabler for them. First, startups have greater flexibility than traditional businesses. They usually do not have excessive layers and cumbersome decision-making procedures and can adapt to market changes and customer needs more quickly. This agility allows startups to launch new products and services faster and flexibly adjust their strategies. Secondly, start-ups are usually more innovative. Start-ups often face limited budgets and tight time constraints. Even larger industry players may be competing for larger industry participants.
- AI 1133 2024-04-22 17:49:01
-
- Four recommended AI-assisted programming tools
- This AI-assisted programming tool has unearthed a large number of useful AI-assisted programming tools in this stage of rapid AI development. AI-assisted programming tools can improve development efficiency, improve code quality, and reduce bug rates. They are important assistants in the modern software development process. Today Dayao will share with you 4 AI-assisted programming tools (and all support C# language). I hope it will be helpful to everyone. https://github.com/YSGStudyHards/DotNetGuide1.GitHubCopilotGitHubCopilot is an AI coding assistant that helps you write code faster and with less effort, so you can focus more on problem solving and collaboration. Git
- AI 1675 2024-04-22 17:34:12
-
- Transformer could be thinking ahead, but just doesn't do it
- Will the language model plan future tokens? This paper gives you the answer. "Don't let YannLeCun see it." YannLeCun said it was too late, he had already seen it. The question discussed in the "LeCun Must Read" paper that I will introduce today is: Is Transformer a thoughtful language model? When it performs inference at a certain location, does it anticipate subsequent locations? The conclusion of this study is that Transformer has the ability to do this, but does not do so in practice. We all know that humans think before they speak. Ten years of linguistic research shows that when humans use language, they mentally predict the upcoming language input, words or sentences. Unlike humans, current language models
- AI 575 2024-04-22 17:22:07
-
- CVPR 2024 | Byte proposes a new generation of data set COCONut, which is denser than COCO granular segmentation
- 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 development of artificial intelligence, language models and generative models have achieved a lot of success, and in the process of designing models, the number of parameters of the models is also increasing. For fine-grained understanding tasks, the number of model parameters is also increasing. However, the scale of existing data sets currently
- AI 1280 2024-04-22 16:20:12
-
- CVPR 2024 | AI can also highly restore the flying skirt when dancing. Nanyang Polytechnic proposes a new paradigm for dynamic human body rendering
- 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. In daily activities, people's movements often cause secondary motion of clothes and thus produce different folds of clothes, which requires an understanding of the geometry and movement of the human body and clothes (human body posture and speed dynamics).
- AI 1008 2024-04-22 14:37:01
-
- The Volcano Engine 4K restoration film was selected for the Beijing Film Festival, and related technology won the championship at CVPR2024
- On April 18, the 14th Beijing International Film Festival officially opened. The Hong Kong films "Plan A" and "The Pledge of Allegiance" jointly restored by the China Film Archive, Douyin and Volcano Engine were selected into the "Tribute and Restoration" section and will be screened on April 18th and 19th. 4K version of "Plan A" and "The Pact" movie posters "Plan A" and "The Pact" were released in 1983 and 1978 respectively, more than 40 years ago. In August 2023, the China Film Archive, Douyin and Volcano Engine jointly launched the "Classic Hong Kong Movie Restoration Project" and announced the restoration of 100 Hong Kong movies. The two films screened this time come from the above-mentioned projects. The original filming negatives of the films were scanned in 4K format in Hong Kong. The entire restoration process included the physical and digital records of the China Film Archive.
- AI 740 2024-04-22 14:10:35
-
- How predictive AI will help achieve net-zero emissions
- Predictive artificial intelligence (AI), a cousin of productive artificial intelligence, uses patterns in historical data to predict future outcomes or classify future events. Experts say the technology can be used to provide actionable insights and aid decision-making and strategy development. Predictive AI leverages large-scale data analytics and machine learning algorithms to discover hidden patterns and trends in historical data and apply them to future scenarios. By understanding past patterns, we can better understand what may happen in the future and strategize accordingly. Predictive AI has applications in various fields, for example over the past year or so we have seen many new and exciting applications of predictive AI emerge in the energy industry to better maintain and optimize energy resource property.
- AI 441 2024-04-22 12:10:01