Shulou(Shulou.com)11/24 Report--

500 questions, the human level that has received higher education, and more emphasis on the real value to industry and life, are designed strictly according to the four dimensions of basic ability index, IQ index, EQ index and tool efficiency index. On August 12, the China Enterprise Development Research Center of Xinhua News Agency Research Institute released the artificial Intelligence Big Model experience report 2.0 (hereinafter referred to as "the report"). IFLYTEK Spark ranked first in the evaluation list of domestic mainstream big models with a total score of 1013. It won the first place in the two dimensions of IQ index and tool efficiency index among the four major evaluation dimensions. The "report" believes that iFLYTEK Spark has "obvious advantages in terms of work efficiency".

According to the report, the iFLYTEK Spark Cognitive large Model, which has seven core competencies (namely, text generation, language understanding, knowledge question and answer, logical reasoning, mathematical ability, code ability, and multimodal ability) has cross-domain knowledge and language comprehension ability. able to understand and perform tasks based on natural dialogue. It continues to evolve from massive data and large-scale knowledge to realize the closed loop of the whole process from proposing, planning to solving problems.

In the face of the "emergence" of domestic large models, iFLYTEK spark cognitive model from project establishment to release, and then to iterative upgrading, every node is in full swing: on May 6, iFLYTEK Spark Cognitive Model and its application in education, office, automobile, human-computer interaction, medical and other fields landed; June 9 broke through the open Q & An and iteratively upgraded in mathematical ability. In the coming August 15, we will achieve a major breakthrough in code capabilities and upgrade in multimodal interaction capabilities; on October 24, we will mark ChatGPT to achieve the same level as Chinese and English.

In terms of IQ assessment, the report believes that humans still have a clear advantage in IQ. The research group considered the large model in terms of common sense knowledge (20%), logic ability (50%) and professional knowledge (30%). The results showed that iFLYTEK ranked first.

Code capability is undoubtedly the underlying capability of iFLYTEK Spark's "high IQ". On August 15, Feixinghuo Code capability will also be greatly upgraded, and relevant code products will be launched at the same time, which is worth looking forward to.

In terms of work efficiency improvement, the "report" believes that AI provides strong support to human beings, and the processing speed of AI is much faster than that of human beings. However, although AI has the advantages of high speed and high efficiency, human wisdom and imagination still play an irreplaceable role in some complex and innovative tasks. The research group focuses on tool efficiency (50%) and generating innovation (50%). The results show that iFLYTEK Spark ranks first and is far ahead with 350 points.

According to the report, compared with June 2023, the current large model products in China have made remarkable progress. However, compared with human beings who have received higher education, the big model still has a certain gap in IQ, EQ and so on. Although AI and human beings show different advantages and disadvantages in different fields, on the whole, the development of AI model has brought an important positive impact on the improvement of human work and life, and the large model is accelerating into life and industry.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.