Reprint public number | China Computer Federation
In this session, senior technical personnel from leading companies such as 360, technical leaders from knowledge graph startups such as Haiyizhi, and top scholars from Fudan University, Southeast University and other universities were invited to discuss in depth the application of knowledge engineering in universities. Opportunities and challenges faced in the language model era, and further understanding of technological breakthroughs, as well as corresponding solutions and cases. April 4th, welcome to sign up!
Provide engineers with a top-level communication platform
CCF TF Issue 97
Time April 4, 2023 19:00-21:10
Knowledge Engineering in the Era of Topic Large Language Models
Welcome to scan the QR code for details and register for the conference
Registration link: https://conf.ccf.org.cn/TF97
With the launch of ChatGPT and other large language models (Large Language Model, LLM), the intelligence of AI, especially the ability to understand and generate expressions, has reached an unprecedented height, and it has also aroused heated discussions in various industries. Various LLM empowerment The discussion and implementation attempts are also different. Therefore, it is necessary for us to re-examine the value of knowledge graphs, especially what kind of knowledge we need, what are the advantages of using graphs for knowledge organization compared to LLM, and how the two complement and enhance each other, all of which are worthy of discussion.
In this context, this conference is honored to invite a number of scholars and technical experts who have been deeply involved in the knowledge graph for many years. At the same time, several speakers have also had in-depth thinking and practice under the recent wave of LLM, and together they will focus on knowledge in the LLM era. On the topic of graph research and new engineering paradigms, we will share the typical combination mode of large models and knowledge graphs, new paradigms of multimodal knowledge engineering research, and the practice of knowledge-enhanced large language models and other important topics. This event aims to build a high-level multi-communication platform, and provide reference and reference for knowledge map builders under the big language model and a wider range of LLM+KG application developers from different levels.
Meeting schedule
TF97 : Knowledge Engineering in the Era of Big Language Models Moderator: Wang Haofen, Chairman of CCF TF Knowledge Graph SIG |
||
time |
theme |
Speaker |
19:00-19:10 |
Event introduction and speech |
Wang Haofen |
19:10-19:35 |
" Talking about the combination of large models and knowledge graphs: some recent direction exploration and summary of experience " |
Liu Huanyong 360 AI Research Institute Algorithm Expert |
19:35-20:00 |
" Thinking and Prospect of Multimodal Knowledge Engineering in AIGC Era " |
Li Zhixu Fudan University Researcher, Doctoral Supervisor |
20:00-20:25 |
" A New Paradigm for AI Application Development Based on KG+LLM " |
Hu Fanghuai Haiyizhi Information Technology (Nanjing) Co., Ltd. |
20:25-20:50 |
" Thoughts on the integration of knowledge-enhanced large language model technology and dual-knowledge platforms " |
Qi Guilin Director of Institute of Cognitive Intelligence, Southeast University |
20:50-21:05 |
Participants Q&A |
Liu Huanyong, Li Zhixu, Hu Fanghuai, Qi Guilin |
21:05-21:10 |
Activity summary |
Ba Chuan Chair of CCF Data Science SIG |
Affiliation SIG
CCF TF Knowledge Graph & Data Science
guest speaker
Liu Huanyong
360 AI Research Institute Algorithm Expert
Topic: "Talking about the combination of large models and knowledge graphs: some recent direction exploration and summary of experience"
Topic introduction: The launch of the large model represented by ChatGPT has set off a new paradigm of NLP and knowledge map implementation. This report will be based on the recent work in the research and development of large models and the direction of knowledge graphs, from the basic data construction of large model research and development, the direction of combining knowledge graphs under the background of large models, and some of the actual landing scenarios of large models and knowledge graphs in 360 The three parts of practice will be introduced, and I will talk about my superficial thoughts and experiences for everyone to think about together.
Personal brief introduction: Algorithm expert of 360 Institute of Artificial Intelligence, person in charge of knowledge map direction algorithm, once worked in the Institute of Software, Chinese Academy of Sciences. The main research direction is the construction and application of domain knowledge/event graphs. He presides over the development of industry-wide event graphs, 360 encyclopedia graphs, knowledge graph platforms, event intelligence analysis, and recommendation on the right side. He has applied for more than ten invention patents and the number of papers articles. In recent years, it has won a number of first and second places in OGB-Wikikg2, CCKS multimodal entity alignment, and interpretable similar case matching. Committed to open source sharing of natural language processing technology, with more than 60 open source projects on github, with more than 20,000 stars, and the establishment of the "Lao Liu Shuo NLP" technology official account, which has extensive influence.
Li Zhixu
Fudan University Researcher, Doctoral Supervisor
Topic: "Thinking and Prospect of Multimodal Knowledge Engineering in AIGC Era"
Topic introduction: AIGC technology brings artificial intelligence into a new era. In the field of multimodal intelligence, AIGC's capabilities continue to improve. Where should multimodal knowledge engineering go? Is it still valuable? How should it develop? In this sharing, the speaker will explore the "dark side" behind the dazzling "light" of the current AIGC technology, and think about and look forward to the multimodal knowledge engineering research in the AIGC era.
Personal profile: Fudan University researcher, doctoral supervisor, assistant director of Shanghai Key Laboratory of Data Science, executive deputy director of Fudan University Knowledge Factory Laboratory, once served as vice president of iFLYTEK Suzhou Research Institute, PhD graduated from University of Queensland, Australia, The main research directions are cognitive intelligence and knowledge engineering, multimodal knowledge graph, big data analysis and mining, etc. He has published more than 150 papers in mainstream journals and international conferences, and presided over more than ten national, provincial and ministerial scientific research projects.
Hu Fanghuai
Haiyizhi Information Technology (Nanjing) Co., Ltd.
Topic: "A New Paradigm for AI Application Development Based on KG+LLM"
Topic introduction: KG is a master of knowledge engineering in the era of big data. With its powerful semantic expression ability, storage ability and reasoning ability, it provides an effective solution for data knowledge organization and intelligent application in the Internet age. With the rapid development of AIGC and LLM-related technologies, LLM represented by ChatGPT has brought a new application interaction paradigm change, which has shown strong natural language understanding and generation capabilities. This report explores the organic combination of KG and LLM, fully combines KG's knowledge representation ability and knowledge reasoning ability, and LLM's language understanding ability and interaction mode, and jointly realizes a new paradigm for the implementation of a new round of AI applications.
Personal brief introduction: Doctor of Natural Language Processing Laboratory of East China University of Science and Technology. 15 years of experience in knowledge map research and industrialization, key members of several national-level projects, published many papers in international conferences such as ISWC and CCKS2017, and top journals, and has been invited to participate in top industry exchange reports many times, including large-scale storage combat analysis, large-scale Knowledge map application, CCKS2017 knowledge map practical report, etc. Proposer of industry knowledge map full life cycle theory, consultant of BIT Big Data Learning Center. He once released the SSCO and Zhishi.me general knowledge graphs based on the data of the three major encyclopedias; for the first time, he proposed the life cycle of industry knowledge graphs and gradually became the industry standard. Participated in the compilation of "2018 Knowledge Graph Development Report", "Knowledge Graph Method, Practice and Application", "Information Technology Artificial Intelligence Knowledge Graph Technical Framework" and other standards and works.
Qi Guilin
Director of Institute of Cognitive Intelligence, Southeast University
Topic: "Thoughts on the Integration of Knowledge Enhanced Big Language Model Technology and Dual Knowledge Platforms"
Topic introduction: This report first introduces knowledge-enhanced large-scale model-related technologies, and on this basis, discusses some thoughts on how to integrate knowledge graph platforms and large-scale model platforms.
Personal brief introduction: Professor of the School of Computer and Software Engineering, Southeast University, director of the Institute of Cognitive Intelligence of Southeast University, chief scientist of Nanjing Keji Data Technology Co., Ltd., one of the founders of OpenKG, deputy director of the Language and Knowledge Computing Committee of the Chinese Information Society of China , Deputy Director of the Knowledge Organization Professional Committee of the Chinese Society for Scientific and Technical Information, Deputy Director of the Knowledge Engineering and Intelligent Service Committee of the Jiangsu Artificial Intelligence Society, Executive Editor-in-Chief of the International Journal of Data Intelligence, Deputy Editor-in-Chief of the International Journal of Web Semantics and Semantic The editorial board member of Web Journal, the editorial board member of Journal of Big Data Research, the data management advisory committee consultant of Elsevier (Elsevier), one of the world's three major publishing groups, and the editorial board member of Information Engineering Journal. Received funding from "Six Talent Peaks of Jiangsu Province" and "Nanjing High-level Talents for Entrepreneurship". He has written 2 monographs and published more than 200 high-level academic papers. Obtained 12 authorized invention patents. Presided over and participated in more than 10 national-level knowledge map-related projects such as key research and development projects of the Ministry of Science and Technology, National 863, National Natural Science Foundation of China, and Natural Science Foundation of China, as well as corporate knowledge map-related projects such as Huawei and Baidu. A paper published by a student under his guidance won the International Conference ICTAI2015 Best Student Paper Award and CIKM Best Short Paper Competition Award. The scientific research achievements obtained have been practically applied in the industry, and have produced practical benefits. They have been implemented in judicial case push, e-commerce data analysis, power failure intelligent detection and knowledge push, medical knowledge question and answer, military decision-making system and security decision-making system. , The prospect of industrialization is broad.
Chairman of SIG
Wang Haofen
Chairman of CCF TF Knowledge Graph SIG, Distinguished Researcher of Tongji University Hundred Talents Program
Personal brief introduction: Tongji University Hundred Talents Program, Distinguished Researcher, Doctoral Supervisor. One of the founders of OpenKG, the world's largest Chinese open knowledge graph alliance. Responsible for participating in a number of national-level AI-related projects, published more than 100 high-level papers in the field of AI, with more than 2,600 citations and an H-index of 25. He built the world's first interactive virtual idol - "Amber Xuyan"; the intelligent customer service robot he built has served more than 1 billion users. Currently, he serves as the deputy director of the Terminology Working Committee of the China Computer Federation, the chairman of SIGKG, the secretary-general of Shanghai, the director of the Chinese Information Society of China, the deputy secretary-general of the Language and Knowledge Computing Committee, and the deputy director of the Natural Language Processing Committee of the Shanghai Computer Federation. Social positions such as secretary-general of Shanghai Jiao Tong University AI Alumni Association.
Ba Chuan
Chairman of CCF TF Data Science SIG, Chief Data Scientist of Athletic World
Personal brief introduction: He once worked for Internet companies such as China Search and Sohu Changyou. The main research areas include data mining, knowledge graph, artificial intelligence, social network, risk control system, recommendation system, data visualization, etc. Deputy director of the Expert Committee of China Education Innovation School-Enterprise Alliance, expert of the Digital Economy Professional Committee of the National Technology Standard Innovation Base (Guizhou Big Data), part-time master tutor of Beihang University, teaching expert of the Graduate School of Xi'an Jiaotong University, part-time teacher and innovation and entrepreneurship mentor of many universities, and many Technology summit speaker and producer.
Knowledge Graph SIG/Data Science SIG event schedule, welcome to follow and participate:
TF97 |
April 4th _ _ |
Knowledge Graph / Data Science SIG |
Knowledge Engineering in the Era of Big Language Models |
TF99 |
April 11th _ _ |
Knowledge Graph SIG |
Construction and application of multi-modal knowledge graph in AIGC era |
TF100 |
April 22nd _ _ |
Knowledge Graph / Data Science SIG |
Data-Centric AI : Knowledge-Driven Data Synthesis |
TF101 |
April 30th _ _ |
Data Science SIG |
Data Science for Global Marketing |
Participation instructions
1. If you are unable to participate after registration, please send an email to apply for cancellation (contact email: [email protected]) before the start of the event . Absence without reason will affect participation in the next event.
2. The event adopts an online mode: Tencent Conference. On the mobile terminal, you can search for "Tencent Conference" in the WeChat applet to log in to the conference, or download the "Tencent Conference" app to log in. For the client, please search for "Tencent Conference" to download and log in.
3. The conference link and password will be notified by email or text message on the day of the event (please check your email for registration after 15:00 on the day of the event). You can click on the Tencent meeting link and enter the password to participate.
4. Please complete the registration before 16:00 on the day of the event , and get the meeting link in time.
5. Free for CCF members, 99 yuan/time for non-members.
member benefits
Members can participate in 20 online events for free, and preferentially participate in 13 offline events, making a good investment for their own technological growth, and an excellent way to acquire professional knowledge with high cost performance!
Professional member/senior member/distinguished member/fellow: 360 yuan/year
Student membership: 50 yuan/year. For specific benefits, please click to view: CCF individual member benefits
If you apply for corporate membership, you can enjoy more free quota, brand promotion and other rights and interests. For details, click to view: CCF corporate membership rights or consultation telephone 0512-65900856 ext. 27
Long press to identify or scan code to join
Participation method
April 4, 2023 19:00-21:00
Long press to identify or scan code to sign up
Registration link:
https://conf.ccf.org.cn/TF97
Contact information
E-mail: [email protected]
Tel: 0512-6590 0856 ext. 27
Mobile: 18912616058
Co
OpenKG
OpenKG (Chinese Open Knowledge Graph) aims to promote the openness, interconnection and crowdsourcing of knowledge graph data with Chinese as the core, and promote the open source and open source of knowledge graph algorithms, tools and platforms.
Click to read the original text and enter the OpenKG website.