Information

On September 19, Mi announced today that Xiaomi-MiMo-Audio, the first original-to-end voice model of open source, had for the first time achieved a generalization of ICL-based samples in the field of voice. According to Mi, five years ago GPT-3 demonstrated for the first time the ability to acquire In-Context Learning (ICL, context learning) through a self-returning language model + large-scale unmarked data training, while in the area of voice, the existing large-scale model still relies heavily on large-scale tagging data and is difficult to adapt to new assignments to human intelligence. And Xia...

According to March 12, Connect, the Young Wida project has invested $26 billion over the next five years to develop an open weight AI model and to build a complete system covering model development, computing infrastructure, talent and ecology. According to the report, the route chosen by Ingweida is not entirely open-source, but rather a "open weight" strategy, i.e. open model parameters, without having to fully comply with open-source agreements. This model is between the closed system of OpenAI and the full open source of Meta Llama, and is more consistent with the need for transparency and customization. In...

Several former and current employees of OpenAI and Google DeepMind recently issued a joint open letter expressing concern about the potential risks of advanced AI and the current lack of regulation of AI tech companies. Image source Pixabay The open letter points out that the development of AI could pose a range of risks, such as exacerbating inequality in existing societies, contributing to manipulation and the spread of disinformation, and the possibility that out-of-control autonomous AI systems could lead to the extinction of the human race. The letter reads that AI companies have powerful financial interests driving them to continue to advance AI research and development, while ......

According to news from April 14, the Humanoid Hub and Forbes reported that yesterday the R1 series of human robots under the banner of Woo Tree Technologies officially opened up global prepurchases, covering North American, European, Japanese, Singapore and Chinese markets, and that official direct-camp shops were online. The R1 series consists of three sections. The introductory version of R1 AIR started at US$ 6,806, the standard version of R1 sold at US$ 8,126, and the flagship version of R1 EDU has not yet been published to contact the sales request. AliEx...

August 6, 2011 - OpenAI announced the release of two open source models, GPT-oss-120b and GPT-oss-20b, the first open source language models to be released by OpenAI since the 2020 release of GPT-2, both of which can be downloaded for free from the Hugging Face online developer platform. Both models, which are available for free download on the Hugging Face online developer platform, have been described by OpenAI as performing "at the cutting edge" in a number of benchmarks used to compare open source models. According to OpenAI, the Gpt-oss-120b model has been shown to perform "at the cutting edge" of core reasoning at .......

KD Xunfei in the "Shenzhen Stock Exchange Interactive Easy" platform to reply to the netizen's question, revealed that the Xunfei Starfire V4.0 against the current capabilities of the GPT-4 is in training, is expected to be released by the end of June 2024, against the GPT-4o products are under development. KU Xunfei said that the company will closely integrate the evolution and breakthroughs in artificial intelligence, and continue to apply the results to each product of intelligent education, and continuously improve the user experience and customer penetration rate of the product. And in April this year, Liu Qingfeng, chairman of KU Xunfei, revealed that on June 27 this year, Xunfei Starfire V4 ......

Baidu Apollo held Apollo Day 2024 in Wuhan Baidu Radish Express Auto Robotics Zhixing Valley, releasing the world's first large model Apollo ADFM (Autonomous Driving Foundation Model) that supports L4-level autonomous driving. According to Baidu, Apollo ADFM reconstructs automatic driving based on large model technology, which can take into account the safety and generalizability of the technology, achieve safety more than 10 times higher than that of human drivers, and achieve city-level coverage of complex scenes in the whole area. Rely on ......

The opening of the World Internet Conference Wuzhen Summit, Alibaba Group CEO Wu Yongming pointed out in his plenary speech that AI promotes productivity changes in thousands of industries, and the value it creates will far exceed the Internet link. Facing the AI era, Alibaba focuses on perfecting the AI infrastructure and adheres to open source and openness to transform AI capabilities into productivity. In the past year, Alibaba has invested heavily in AI infrastructure, and because of the scale effect and technological advances, the cost of computing power has been reduced, and the price of "Tongyi Thousand Questions" API call has dropped by 97% in a year, and the cost of calling one million tokens has been as low as 0.5 yuan. He believes that cheap arithmetic will take us into the era of intelligence ......