Following the successful development of InvestLM - Hong Kong's first open-source large language model (LLM) for financial generative AI (GenAI) applications, the School of Business and Management of the Hong Kong University of Science and Technology (HKUST Business School) has just launched the InvestLM Generative AI Platform, which will be opened for free registration to the financial services industry in phases. The platform is capable of handling finance-related GenAI tasks and generating human-like responses comparable to those of well-known general-purpose chatbots.
The InvestLM Generative AI Platform is designed to support the local financial services industry, particularly small- and medium-sized financial firms, in harnessing the potential of GenAI technology. Effective immediately, local financial institutions may visit the platform's website (investlm.hkust.edu.hk)1 and register as free users. The initial phase will offer a limited number of user accounts, considering potential user demand and system capacity2.
InvestLM has been optimized for analyzing and generating financial text. This enhancement aims to improve the efficiency of finance and investment professionals in tasks such as summarizing financial news and reports, conducting market sentiment and topic analysis, answering financial tabular questions, performing ESG (environmental, social, and governance) analysis and more3.
Prof. TAM Kar-Yan, Vice-President for Administration and Business and Chair Professor of Information Systems at HKUST, said, "AI application can potentially enhance the efficiency of the financial services industry. The new InvestLM Generative AI Platform reflects HKUST's commitment to promoting the wider use of GenAI in the financial services industry, particularly the small- and medium-sized enterprises, many of whom are resource-constrained to undertake related service development. The platform will provide an educational and testing venue for those enterprises to explore and evaluate the potential of generative AI solutions."
Prof. Tam added, "Running on HKUST's computing resources, the InvestLM Generative AI Platform builds on the Government's call for the industry to make full use of available AI models and infrastructures in the market. By taking a phased approach in offering this free service, we hope to ensure optimal user experience and system performance. We will closely monitor user demand to determine how widely we can make the service available to the industry."
Prof. YANG Yi, Associate Professor of Information Systems at HKUST, and a member of the InvestLM research project team, elaborated, "InvestLM's responses have been rated as comparable to those of state-of-the-art general-purpose LLMs by financial experts, such as hedge fund managers and analysts. We have specifically tailored the model for finance and investment GenAI tasks, mitigating over-alignment issues associated with general-purpose models, and making InvestLM more aligned with economic rationality. We welcome feedback from users to improve the platform and explore additional use cases that promote responsible AI development in Hong Kong's financial services industry."
The InvestLM research team appreciates the support and collaboration of various public and private sector organizations in launching this platform within the financial services industry. Among the various collaborations, engagement sessions will be co-hosted with the Hong Kong Monetary Authority to familiarize the banking sector with the InvestLM Generative AI Platform.
Notes:
1. As the platform relies on finite computer resources, its response speed is subject to various factors, including the number of concurrent users and network traffic. As an emerging and evolving technology, GenAI's reliability and accuracy cannot always be assured, and there is a possibility that it may generate unfounded answers. Therefore, users are reminded to fact-check and exercise their judgment when using the platform.
2. If the number of registrations exceeds the available user accounts during this initial phase, the InvestLM team will apply criteria to select the successful registrations with a view to ensuring a broad range of financial institutions can try this online service.
3. The HKUST Business School research team initially developed InvestLM in 2023 by adapting from the open-source general-purpose LLM known as LLaMA1. The team has since updated the model by adapting from LLaMA3.1, an open-source general-purpose LLM, with a high-quality and diverse set of finance and investment-related generative AI tasks. InvestLM Generative AI Platform is powered by this updated version of InvestLM, which has about 70 billion parameters.
About The Hong Kong University of Science and Technology
The Hong Kong University of Science and Technology (HKUST) (https://hkust.edu.hk/) is a world-class university committed to innovative education, research excellence, and impactful knowledge transfer. Our holistic and interdisciplinary pedagogy approach is recognized globally. HKUST placed 3rd in the Times Higher Education's Young University Rankings 2024, 19th Worldwide and No.1 in Hong Kong in Times Higher Education's impact Rankings. Twelve of HKUST's subjects were ranked among the world's top 50 in the QS World University Rankings by Subject 2024. Of these, "Data Science and Artificial Intelligence" ranked first in Hong Kong and 10th in the world. Our graduates are highly competitive, consistently placing among the world's top 30 most sought-after employees. Our research and entrepreneurship endeavors are also innovative and impactful. Over 80% of our work was rated "Internationally excellent" or "world leading" in the latest Research Assessment Exercise 2020 of Hong Kong's University Grants Committee. As of October 2024, HKUST members have founded 1,800 active start-ups, including 10 Unicorns and 14 exits (IPO or M&A).
About HKUST Business School
The HKUST Business School is young, innovative and committed to advancing global business knowledge. The School has forged an international reputation for world class education programs and research performance, and has received many top global rankings. For more details about the School, please visit https://bm.hkust.edu.hk.
