Scan the QR code:
(Scan the QR code to register)
With the progress made in the research of large models, the ability of models continues to increase, and the practice of application in the industry is also in full swing. However, there is still a considerable gap between the existing industry applications based on the pedestal model and the credible, reliable and efficient expected by the industry, as well as the real efficiency and influence generated in the industry, and the typical industry application results of the large model are still insufficient. Based on this, the academic community, the providers of large model capabilities, and the model application parties with real industry scenarios and data have all shown great interest in the above problems, and have carried out in-depth thinking and systematic work from their own perspectives.
This forum aims to invite academic experts, pedestal model suppliers, industry model suppliers, and industry application parties with front-line experience in the application of large models in the industry, to discuss the technical practices of system construction, data governance, landing scenario docking, and private model fine-tuning in the actual implementation of large models, analyze the difficulties and dilemmas of large models in industry applications, exchange relevant lessons and lessons, and give full play to the advantages of academia, model capability suppliers, and industrial application parties. Jointly promote the development of large models and provide some valuable views and references.
This forum plans to output a practical tool group for the implementation of large models, and provide a list of challenges to the capabilities of the model itself in various landing scenarios, so as to form a technical paradigm and engineering technology path for the industrial application of large model technology, and promote the landing and application of large models in the industry.
Forum arrangement
Time: 13:30-17:30, May 18
Location: Lanson Place on the 5th floor of Ningbo Nanyuan Hotel Business Building
| Introduction | topic | Keynote speaker | unit |
| 1 | R&D and thinking of large language model technology | Zhao Xin | Chinese People's University |
| 2 | Exploration of the application of GLM pedestal large model | Lee Ji-sung | Zhipu AI Large Model Division |
| 3 | Judicial large-scale model training and application exploration of large-scale models combined with court practice | Luo Cheng | Beijing Mejia Intelligent Technology Co., Ltd |
| 4 | Practical application of artificial intelligence technology innovation of Shandong Energy Group | Xu Jiali | Shandong Energy Group Information Center |
| 5 | AI-Driven Future: Shell in the Housing Industry | Xiao Peng | Shell looking for a house |
Executive Chairman
Zhao Kai
CCF YOCSEF AC委员
Associate Researcher, Institute of Automation, Chinese Academy of Sciences
Doctor of Engineering, research and technology in the field of decision-making intelligence and autonomous operation robots, the main work is committed to the combination of high-performance artificial intelligence and robot control systems, research and development of highly autonomous, highly intelligent industry landing robot ontology and systems. He has participated in a number of research and R&D projects such as the major project of the Ministry of Science and Technology of the People's Republic of China (2030), the National Natural Science Foundation of China, the key R&D plan of Jiangsu Province (industrial prospect and key core technologies), the transformation and upgrading of industry and information industry in Jiangsu Province (engineering research on key core technologies and equipment), system and algorithm development, etc. He has published 9 academic papers in journals or conferences, of which 1 EI conference paper as the first author won the Best Paper Award of the 14th International Conference on Communication Technology ICCT2012; 4 national invention patents have been authorized. Responsible for or as a core member to complete a number of horizontal development projects entrusted by the enterprise.
Co-Executive Chairman
ZHANG Ying
CCF YOCSEF AC委员
Professor, North China Electric Power University
Ph.D. in Engineering, research direction is artificial intelligence application research, mainly used in smart grid, intelligent transportation, civil engineering and medical fields. HE HAS PUBLISHED MORE THAN 50 ACADEMIC PAPERS, INCLUDING TOP JOURNALS IN THE FIELDS OF IEEE TRANSACTIONS IEEE T SMART GRID, IEEE T INTELL TRANSP, AND CCF CLASS A CONFERENCE IJCAI. It has obtained 11 authorized national invention patents, published 3 academic monographs and 2 academic translations. He has won 1 Beijing "Young Talent" Program, presided over 2 projects of the National Natural Science Foundation of China, presided over 2 provincial and ministerial scientific research projects, and presided over more than 30 horizontal commissioned research projects in the field of artificial intelligence applications. He serves on the editorial boards of several international academic journals. He is currently a member of the CCF Artificial Intelligence and Pattern Recognition Committee (2018-) and a member of the French Institute for Modeling, Analysis and Decision Making in Dynamic Systems (GdR MACS) (2017-).
Forum Speaker
Zhao Xin
CCF Distinguished Member
Professor, Hillhouse School of Artificial Intelligence, Chinese Minmin University
He received his Ph.D. from Peking University in July 2014 and has been working at Chinese University since then. His research field is information retrieval and natural language processing, with more than 100 papers published and more than 17,000 citations by Google Scholar, and he has led the research and development of the Yulan large language model, and organized the compilation of the large language model review paper "A Survey of Large Language Models" (preprint article) and the Chinese book "Large Language Model". He has won the 2020 Wu Wenjun Artificial Intelligence Outstanding Youth Award, the ECIR 2021 Time Test Award, and the CCF-IEEE CS Young Scientist Award.
Title: R&D and Thinking of Large Language Model Technology
Summary:
Recently, large language models represented by ChatGPT have received widespread attention from society. This report will focus on the key technologies in the research and development process of large language models, discuss from the aspects of pre-training, analyze the characteristics and limitations of existing technologies, and give the speaker's relevant thoughts and practical experience in these aspects.
Lee Ji-sung
Vice President of Technology of Zhipu AI Large Model Division
He has successively served as the head of the knowledge question and answer algorithm of Alibaba Artificial Intelligence Lab and the director of AI algorithm of Full Truck Group. His research interests include artificial intelligence, machine learning, and natural language processing. He led the construction of Alibaba's first intelligent hardware device, Tmall Genie's knowledge question and answer engine, and the construction of the AI platform of Full Bang Group. He is the main researcher of XLIKE, the seventh framework international cooperation project of the European Union, the research backbone of three national key R&D programs, and has presided over and completed one national project and three provincial and ministerial projects.
Title: Exploration of GLM Base Large Model Application
Summary:
As a fully self-developed domestic large model, GLM's performance is comparable to the world's leading level. GLM has conducted in-depth exploration in central state-owned enterprises, financial institutions and other units, and has accumulated very rich experience in computing power, data, scenarios and other aspects.
Luo Cheng
Founder and CEO of Beijing Mejia Intelligent Technology Co., Ltd
He received his bachelor's degree and doctorate degree from the Department of Computer Science of Tsinghua University, and was engaged in postdoctoral research at Tsinghua University, and won honors such as the Outstanding Doctoral Dissertation of the Chinese Society of Chinese and Information, the Best Essay of ACM SIGIR, and the Best Paper Award of AIRS. He was a visiting scholar at the National University of Singapore and Waseda University in Japan. He was a visiting researcher at the NExT Centre in Singapore.
Title: Judicial Large Model Training and Exploration of Large Model Application Combined with Court Practice
Summary:
In the past decade, the number of cases accepted by the people's courts has increased by 2.4 times, which has brought a huge workload to the judicial organs. The progress of pre-trained model technology has brought new space for the intellectualization of multiple practical scenarios of judicial organs. This report will focus on the training path of LegalAID, a large model of Tsinghua University, and introduce the exploration and practice of LegalAID in accurate case retrieval, document writing and review, and litigation-related mediation for element matching.
Xu Jiali
Director of Information Center of Shandong Energy Group
Doctor of Engineering, Engineering Technology Application Researcher, Vice Dean of Industrial Research Institute of Shandong University. Since joining Shandong Energy Group from Shandong University in 2013, he has been engaged in the research and management of coal mine intelligence, automation and informatization, and has won the first prize of Shandong Science and Technology Progress Award and China Coal Youth Science and Technology Award.
Title: Practical Application of Artificial Intelligence Technology Innovation of Shandong Energy Group
Summary:
Shandong Energy Group and Huawei have been working together since 2022 to develop the large model of mines, and after about two years of hard work, they have completed the research and development of CV models and prediction models in the mining field, and released the industry's first commercial mine model on July 18, 2023, forming a solution covering 9 industries and 63 scenarios, and applying it on a large scale in more than 30 factories and mines.
Xiao Peng
Technical Director of Beike Housing and Head of Beike Infrastructure
He has worked in Sino-i and Weibo. He has been engaged in cloud computing, big data and machine learning related infrastructure construction for a long time, focusing on efficiency improvement and FinOps related fields, and has a deep understanding of the residential industry.
Title: AI-Driven Future: A Case Study of Beike in the Housing Industry
Summary:
As a deep cultivator in the housing industry, Beike has been continuously exploring and practicing in the wave of AI technology innovation. With the rise of AI large-scale model technology, we are actively building a way to integrate AI with the housing industry, and gain insight into its future development direction. This report aims to introduce in detail how Beike applies AI large model technology to the residential industry, and to discuss in depth the strategies, challenges and solutions in its application. At the same time, we will share our insights and forward-looking outlook on the future development trends of the residential industry, aiming to reveal how AI is shaping the new future of the residential industry.