CCF Technology Achievement Awards Recognize Ant Group’s Advances in AI‑Driven Financial Risk Modeling and Edge Intelligence for Alipay

On October 25, 2024, the China Computer Conference (CNCC) presented the annual CCF Technology Achievement Awards, honoring two Ant Group projects with the CCF Technology Progress Second Prize and Third Prize.

The award‑winning project “Key Technologies and Applications for Complex Behavior Modeling in Digital Inclusive Finance Security” was a joint effort between the Chinese Academy of Sciences’ Institute of Computing Technology and Ant Group, recognized for its deep‑sequence modeling, graph‑machine learning, and adversarial training methods that analyze multi‑source heterogeneous behavior data to enhance financial risk control and significantly reduce fraud losses.

Since 2018 the project has produced 22 papers (including 13 at CCF A‑class conferences), received best short‑paper nominations at SIGIR 2021 and CIKM 2022, and secured 20 national invention patents.

The second award‑winning project “Alipay Terminal Intelligent Technology and Application” earned the CCF 2024 Technology Progress Third Prize for its edge‑AI solutions that enable on‑device inference with real‑time response, strong privacy protection, and low cost, addressing challenges of large model sizes, memory usage, and diverse hardware across thousands of smartphones and IoT devices.

Leveraging the self‑developed xNN inference engine, the team achieved model and parameter compression, resource optimization, and inference acceleration, attaining top rankings in ICCV‑2023 and CVPR‑2023 challenges and contributing to the “Gemini” edge‑cloud collaborative AI framework selected by the Ministry of Industry and Information Technology.

These technologies have been deployed at massive scale in Alipay, supporting billions of users and millions of merchants, and have supported major national projects such as the “Asian Games One‑Stop” initiative.