Affinit Wins Grand Prize at ACM ICAIF 2025 AI Finance Conference
AI financial innovation company Affinit (formerly Balance Hero) achieved remarkable results and proved its global technological prowess at 'ACM ICAIF 2025 (ACM International Conference on AI in Finance 2025),' hosted by ACM (Association for...
AI financial innovation company Affinit (formerly Balance Hero) achieved remarkable results and proved its global technological prowess at 'ACM ICAIF 2025 (ACM International Conference on AI in Finance 2025),' hosted by ACM (Association for Computing Machinery), the world's most prestigious academic society. At the recent awards ceremony held in Singapore, Affinit remarkably took 1st and 5th places in two competitive categories, respectively.
Affinit's research team of Shin Jae-hyuk and Jeon Hyun-woo took the honorable 1st place in 'FinSurvival Challenge - Deep Survival Modeling for Financial Transactions,' a key competitive category of the conference. In this category, Affinit, through its unique framework and multi-level indexing strategy, innovatively solved the problems of incomplete data and data leakage inherent in financial transaction records. In particular, by effectively applying the XGBoost AFT (Accelerated Failure Time) survival analysis model and a two-stage learning strategy, it achieved outstanding predictive performance, receiving high praise from the judges.
Furthermore, in the ‘FinDDR 2025 - Financial Document Deep Research Challenge’ category, Affinit proudly secured 5th place, surpassing leading global institutions such as Fudan University Shanghai Innovation Institute (1st), Microsoft (3rd), and global credit rating agency Experian (4th), once again demonstrating its AI technology prowess. This achievement is thanks to the RAG (Retrieval-Augmented Generation)-based AI agent technology independently developed by Affinit's Data Science Team. This AI agent demonstrated an innovative ability to automatically chunk and embed vast annual reports, then generate high-quality financial analysis reports.
Based on the RAG and LLM (Large Language Model) technologies validated through this competition, Affinit plans to expand its business into core areas of financial decision-making, such as AI-based credit scoring, financial analysis, and risk prediction.
Shin Jae-hyuk, Head of AI, stated, “This achievement clearly confirms that Affinit’s LLM technology has global competitiveness,” and particularly emphasized the significance of achieving the goal of generating research reports through automatic analysis of multilingual annual reports. Lee Chul-won, CEO, said, “The overwhelming 1st place in our core competency area and a top-tier entry in a new challenging area are proof that Affinit is ready to lead the financial AI market,” adding, “We will continue to open new horizons in the financial AI market through continuous technological innovation.”
What's Your Reaction?
Like
0
Dislike
0
Love
0
Funny
0
Angry
0
Sad
0
Wow
0