nearly 5,000 teams competed for the top spot in financial ai, and the afac2024 competition concluded successfully at the bund
2024-09-21
한어Русский языкEnglishFrançaisIndonesianSanskrit日本語DeutschPortuguêsΕλληνικάespañolItalianoSuomalainenLatina
since chatgpt was launched in december 2022, a global ai competition has been launched, and many countries are going all out to seize the commanding heights. in order to better tap, discover and cultivate talents, this year's 2024 inclusion·bund conference will set up an ai innovation competition for the first time, aiming to provide young scientific and technological talents with an exclusive platform to showcase their innovative achievements and demonstrate the infinite vitality of scientific and technological innovation. the "science and technology intelligence innovation competition" is guided by the shanghai science and technology commission and includes two important events: the "afac2024 financial intelligence innovation competition" and the "global deepfake attack and defense challenge".
under the guidance of shanghai science and technology commission, supported by china computer federation (ccf) and shanghai science and technology entrepreneurship center, more than 20 famous universities and enterprises including peking university, fudan university, university of hong kong, and ant group jointly launched the afac2024 financial intelligence innovation competition (hereinafter referred to as "afac competition"). on september 6, at the bund conference, the afac competition roadshow and award ceremony came to a successful conclusion.
from the official launch of the competition on may 30 to the final decision at the end of august, nearly 5,000 technical teams explored the application prospects of ai in the financial field from their own unique perspectives. sparks of innovation emerged at the bund conference, nurturing the next opportunity to change the world and inspiring more people to join this endless journey of exploration.
the scene of the award ceremony of the afac competition.
as one of the ai innovation competitions of the bund conference, this year's afac competition has been fully upgraded. it is not only jointly initiated by dozens of famous universities and companies around the world, but also focuses on large-scale model technology and innovative entrepreneurship, aiming to continuously explore the implementation of large-scale model technology in financial technology and promote innovative development. the competition invited a number of world-renowned top universities such as the university of hong kong, peking university, zhejiang university, tongji university, and nanyang technological university business school, as well as well-known financial technology companies such as ping an technology, taikang online, and china merchants bank to serve as members of the organizing committee. nearly 50 industry-university-research elites formed an expert committee to provide professional guidance for the competition. the competition also provides participating teams with supporting services such as a million-dollar prize pool, real industry propositions, massive real industry data, and key industry support.
4882 teams compete on the same stage to determine the champion
this year's competition has been upgraded based on last year's competition. in addition to the challenge group, two new groups, the startup group and the enterprise group, have been added, forming a comprehensive competition system of algorithm competition, application competition and entrepreneurship competition, encouraging participants to continuously explore and make breakthroughs in the field of artificial intelligence algorithms and solve complex technical problems. after fierce competition among 4,882 teams, the winners of the three groups were finally decided.
among them, a total of 11 excellent projects stood out in the enterprise group. the first prize winner of the enterprise group, shenzhen qing technology, used the qiankunquan agent platform to solve the core problems of illusion control of large models in the financial field, business data and knowledge access, compliance and security of generated results, end-to-end efficiency of production, research and operation, and quality assessment of generated results. it integrated the big model capabilities into the "system engineering" of financial business, solved the rigid business needs, and provided impetus for the "digital, intensive and refined" customer management of financial institutions.
at the bund conference, representatives from ningbo shenqing information technology co., ltd. from the corporate group shared their views.
the "yanshengxing ai agent" project, which won the second prize in the enterprise group, is committed to building an integrated intelligent process platform based on ai big model technology, aiming to automate software operations through conversational interaction. the project scope includes creating and managing channels, recording and choreographing workflows, and automating process execution. through multimodal data collection and flexible calling of ai models, the project is committed to achieving comprehensive intelligent transformation of enterprise processes.
the "health insurance ai underwriting assistant", which is also the second prize winning project in the enterprise group, is based on the medical big model and empowers the health insurance underwriting process, assisting traditional manual underwriting, improving underwriting efficiency, reducing underwriting costs, and shortening underwriting time.
at the bund conference, representatives from ping fuyin health technology (shanghai) co., ltd. from the corporate group shared their views.
in the start-up group, the winning project "smart travel fantasy" is an ai-driven smart travel assistant that provides itinerary customization, smart travel companionship, video travel diary generation and other capabilities, covering a variety of travel scenarios. it can customize a detailed travel plan including transportation, hotels, and attractions according to user preferences, budget, etc., and can also generate videos based on the itinerary, allowing users to experience the entire itinerary more intuitively in the form of videos; it provides transportation planning, attraction explanations, etc. during the trip, and provides smart travel companion services anytime and anywhere.
at the bund conference, representatives from the zhiyou fantasy team from the start-up group shared their views.
the challenge group continued the algorithm track of the first competition. this year, there are four major competition topics, including real propositions from ant group's industry practice, and academic research competition topics with fudan university and east china normal university, namely "financial instrument learning", "questions and answers based on insurance terms", "aigc financial multimodal research report intelligent generation", and "contradiction identification and vulnerability discovery in long texts of financial rules". during the competition, hengsheng juyuan provided free market data support throughout the whole process, providing convenience for the contestants. these four major competition topics also determined a group of excellent teams.
huang chao, assistant professor at the school of computer science at the university of hong kong, served as a judge for the challenge group "identification of contradictions and discovery of loopholes in long texts of financial rules". during the review process, he was impressed by the innovative solutions of the contestants. "the time for the contestants is actually quite tight. after the topic is released, the contestants usually only have a few days to complete the solution, but i was surprised that they still completed it to a high degree within the time limit. what's more surprising is that the contestants are generally young this time, and the entry points of the innovative solutions are beyond the judges' expectations. for example, the use of prompt word design and fine-tuning to achieve compliance and error correction in finance, the methods are diverse and have a great possibility of implementation."
huang hailiang, dean of the school of information management and engineering at shanghai university of finance and economics and a judge of the competition's initial group, said that the competition proposals such as research report opinion extraction and financial risk event reasoning are highly consistent with actual financial needs and are close to the industry's practical level.
solve real industry issues and promote the application and popularization of financial technology
from big data credit reporting solutions for building a trusted financial ecosystem to applications that usher in a new era of smart tourism, to the financial industry transformation driven by aigc, to the national legal intelligent advisor built based on the agentuniverse multi-agent framework, and to the ai assistant that leads a new chapter in the intelligence of health insurance, a number of application cases that emerged in the afac competition all demonstrated the powerful potential of ai technology.
the afac2024 financial intelligence innovation competition not only unearthed a number of outstanding projects with market potential and innovative value, but also promoted the widespread application and popularization of financial technology. the competition also strengthened cooperation and exchanges among government departments, investment institutions, incubation parks and other parties, injecting new impetus into the construction of the financial technology innovation ecosystem. as huang hailiang said, in this competition, the contestants had a deep collision with the scientific and academic circles, which helped boost the confidence of the pioneers of big models in the financial industry and promoted industry-university-research cooperation.
li hui, chairman of the executive committee of the afac2024 financial intelligence innovation competition, said that the competition is a connector and bridge for innovation and entrepreneurship. in order to give birth to new formats and models and drive rapid growth and breakthroughs in industries, technologies, and talents, the competition provides winning projects with industry and investment and financing docking services, giving winning companies the opportunity to establish connections with upstream and downstream companies in the industrial chain, investment institutions, government departments, etc., to promote the industrialization and commercialization of projects.
applying outstanding projects to the industry is a proposition that requires continuous iteration and optimization. li hui said that this year's competition has designed a continuous tracking mechanism. on the one hand, it continuously tracks and evaluates the implementation of projects and market performance, and on the other hand, it continuously invites mentors with rich industry experience to provide guidance and suggestions to participating teams during and after the competition. "we are also considering building an expert database covering multiple fields such as financial technology and artificial intelligence. we can build an innovation and entrepreneurship service platform based on the competition or go beyond the competition to promote exchanges and cooperation between academia and industry. while providing innovative impetus for the competition, it promotes the deep integration of the two sides to form good interaction and a virtuous circle."
support outstanding ai projects and open up the channel between industry, academia and research
it is understood that future events will strengthen cooperation with technology and finance alliance companies, technology innovation funds, and incubators to provide more technology innovation carrier services for the transformation of innovative achievements, while helping to deepen the integration and interaction of innovative talents, entrepreneurial projects, and scientific research industries. "we are determined to create conditions for start-up teams and small and medium-sized enterprises in innovation and entrepreneurship," said li hui.
this year, more diverse faces appeared in the afac competition, from 19-year-old freshmen to 42-year-old veteran contestants. among them, there are undergraduate students and senior engineers who have worked in the industry for many years and love algorithms. xia maojin, a 32-year-old algorithm manager, balances family, work and competitions, and is increasingly aware that in the era of large models, traditional data processing and analysis methods can no longer solve real problems encountered in work scenarios. only by trying new ideas and methods can we stand out in the fierce competition.
talking about the significance of the competition to the contestants, li hui said, "some contestants are entering the workplace from campus, taking a solid step in their careers, while others should be experienced technical experts who are trying to find a balance between their busy work and life, and regard the competition as an extension of their technical exploration and passion."
"when exploring cutting-edge ai, vitality is very important. these contestants are in an era full of opportunities. it is very exciting that they can participate in the research of big models during their undergraduate studies." huang chao hopes that the contestants can continue to delve deeper into the field of big models, and looks forward to seeing big models being implemented in more financial technology scenarios in the future.
the afac2024 financial intelligence innovation competition is not only a technological competition, but also a vision about finance and the future. it will accelerate the application of big models in the field of financial technology and allow financial ai services to enter the lives of ordinary people like scanning code payment.
in the future, the afac2024 financial intelligence innovation competition will continue to take "solving real industry issues" as its original intention, adhere to the purpose of "gathering and cultivating outstanding scientific and technological talents, conducting high-level cooperation and exchanges, and creating top financial intelligence events", and take "financial technology, big model technology, innovation and entrepreneurship" as the core axis to incubate talents and innovative projects in the financial field.
thepaper.cn reporter jialiluo
(this article is from the paper. for more original information, please download the "the paper" app)
