With the continuous development of data intelligence technology, content generation technology represented by AIGC driven by large language models (LLM) has become an indispensable part of enterprises’ data intelligence capabilities. However, traditional content generation technologies face issues such as untimely information updates, lack of vertical domain knowledge, and model hallucinations. The Retrieval-Augmented Generation (RAG) technology provides an effective solution to these problems, becoming a major trend in the era of data intelligence.
Retrieval-Augmented Generation is a technical solution that improves content generation effectiveness by retrieving from external knowledge bases. By introducing a retrieval process, it can effectively alleviate the hallucination issues of AIGC technology, speed up knowledge updates, and enhance the diversity and traceability of generated content, making it more practical and trustworthy in real-world applications.
In this salon, we are fortunate to invite experts in the field of Retrieval-Augmented Generation (RAG) to introduce and share practical cases of RAG technology applications. We hope to help all parties deepen their understanding of RAG technology, accelerate the rapid implementation of large models across various industries and enterprises, and discuss the future development trends of RAG technology together. Interested friends are welcome to click the card below to make an appointment:
💠 Salon Agenda:
💠 Detailed Introduction:
Han Xiaolu, Engineer at the Cloud Computing and Big Data Research Institute of China Academy of Information and Communications Technology
Personal Introduction: Engineer at the Cloud Computing and Big Data Research Institute of China Academy of Information and Communications Technology, responsible for research on large model + data intelligence applications, leading the preparation of standards such as the “Data Ethics White Paper” and the “Technical Requirements for Intelligent Data Analysis Tools Driven by Large Models”, “Technical Requirements for Intelligent Knowledge Graphs Driven by Large Models”, “Technical Requirements for Retrieval-Augmented Generation (RAG)”, and “Technical Requirements for Intelligent Q&A Systems Driven by Large Models”.
Speech Topic: Key Technologies for Implementing Large Models in the Era of Data Intelligence: Introduction to the Development Trends of Retrieval-Augmented Generation RAG and Related Standardization Work
Speech Outline:
1. Introduction to the Standardization Work of Data Intelligence Application Directions
2. Introduction to the Development Trends of Retrieval-Augmented Generation (RAG)
3. Release of the “Technical Requirements for Retrieval-Augmented Generation RAG” Standard
Ren Xiang, Product Head of Tencent Cloud Big Data ES
Personal Introduction: Rich experience in both B2B and B2C products, responsible for the construction and commercialization growth of Tencent Cloud ES from 0 to 1, previously engaged in instant messaging, security, community, and other product work, holding more than ten patents.
Speech Title: RAG Standards and Technical Practices of Tencent Cloud ES
Speech Outline:
1. Major Contributions of Tencent Cloud in RAG Standard Development
2. Solutions and Advantages of Tencent Cloud ES in RAG
3. Sharing of RAG Practice Cases of Tencent Cloud ES
Audience Benefits:
1. Understand RAG technology and its application scenarios
2. Understand RAG standards and related categories
3. Understand the characteristics and advantages of ES as a RAG solution
Fan Yixing, Associate Researcher at the Institute of Computing Technology, Chinese Academy of Sciences
Personal Introduction: Fan Yixing, Associate Researcher at the Institute of Computing Technology, Chinese Academy of Sciences. Main research areas include information retrieval and natural language processing, selected for the sixth session of the China Association for Science and Technology Youth Talent Support Project, member of the Youth Innovation Promotion Association of the Chinese Academy of Sciences, published over 40 papers at top international academic conferences such as SIGIR, WWW, CIKM, won the 2017 CIKM Best Paper Runner-Up Award and the 2018 Excellent Doctoral Dissertation Award from the Chinese Information Processing Society. Serving as an executive member of the Information Retrieval Committee of the Chinese Information Processing Society, a member of the Youth Working Committee of the Chinese Information Processing Society, and a committee member of various domestic and international conferences. Developed the deep text matching tool MatchZoo, widely used and recognized by researchers on the open-source platform GitHub, accumulating over 4000 stars, extensively utilized by more than 100 universities and enterprises at home and abroad.
Speech Title: Exploration and Practice of Intelligent Information Assistants Based on Retrieval-Augmented Generation
Audience Benefits:
1. Introduction to RAG challenges
2. Our exploration and practice
Zhou Yuan, Co-founder and Chief Scientist of Guanyuan Data
Personal Introduction: Co-founder of Guanyuan Data, focusing on the research and development of innovative products in the AI + BI direction.
Speech Title: RAG Applications in the Field of Data Analysis
Speech Outline:
1. Challenges in the ChatBI scenario
2. How ChatBI combines with RAG
3. Challenges and responses related to RAG
Audience Benefits:
1. Understand the way ChatBI combines with RAG and the differences from typical Q&A scenarios
2. Reference for RAG technology architecture design and component selection
3. Introduction to the difficulties and experiences of RAG implementation
Xu Songlin, Senior Architect at the Wisdom City Design Institute of Yidian
Personal Introduction: Senior architect, information system project management engineer, data governance expert (CDGP), CCF professional member, research areas include big data, privacy computing, artificial intelligence, etc.
Speech Title: Retrieval Augmentation: A New Path for Empowering Standard Literature Practice
Speech Outline:
1. Background and Current Status: Development background and current status of standard literature in recent years
2. Architecture and Optimization: RAG architecture and optimization in conjunction with standard literature
3. Empowerment and Outlook: Scene empowerment in the government sector and future prospects
Audience Benefits:
1. Relevant policy background and recent developments in standard literature over the past three years
2. How to integrate standard literature and other unstructured documents to achieve retrieval augmentation
3. Understanding common optimization ideas for retrieval augmentation
4. Series of scene empowerment and outlook for large models in the government sector
Li Yongfei, Artificial Intelligence Expert at Longkun Suchang
Personal Introduction: 12 years of software industry experience, deeply involved in industrial intelligence and machine learning fields. Currently focusing on the promotion and implementation of large model industrial applications, dedicated to the exploration and innovation of industrial knowledge.
Speech Title: Practical Implementation of the Yaoguang Large Model in Industrial Scenarios
Speech Outline:
1. The path of innovation in customer business empowered by Longkun AI technology
2. Problems and challenges faced by large models in industrial scenarios
3. How industrial enterprises respond to the opportunities and challenges brought by new technologies
4. Thoughts and understanding of the implementation of large models in the industry
5. Introduction to the architecture of the Suchang Yaoguang large model
6. Introduction to the capabilities of the Suchang Yaoguang large model
7. Introduction to the application scenarios of the Suchang Yaoguang large model
8. Introduction to typical cases of the Suchang Yaoguang large model
Audience Benefits:
1. How to improve innovation efficiency through large models?
2. How can industrial enterprises implement large models?
3. How to achieve digital transformation through large models?
Click the card to make an appointment for the live broadcast