Congratulations to the team from A*STAR Centre for Frontier AI Research (A*STAR CFAR) on being nominated the Best Paper Award at the Association for Computing Machinery’s International Conference on Multimedia 2024 (ACM MM 2024).

Held from 28 Oct – 1 Nov 2024 in Melbourne, Australia, ACM MM is one of the top-tier global forums dedicated to multimedia research, bringing together leading researchers, developers, and practitioners to explore the latest advances in multimedia technology and its applications.

The team’s paper, titled “Evolution-aware Variance (EVA) Coreset Selection for Medical Image Classification” is one of only 26 papers chosen from 4,325 accepted papers to receive this prestigious award.

The researchers contributing to this research include:

• Dr Joey Zhou, Deputy Director and Principal Scientist
• Dr Zhang Xin, Scientist
• Ms Hong Yuxin, Intern

The paper delves further into the Temporal Dual-Depth Scoring (TDDS) method introduced in a previous study by Dr Zhang Xin, advancing dataset compression strategies for high-dimensional medical imaging data.

The proposed method, Evolution-aware Variance (EVA), captures the evolutionary process of model training through a dual-window approach, reflecting the fluctuation of sample importance via variance measurement. This innovation enhances intra-class similarity and inter-class discrimination, making it particularly effective at higher compression ratios. EVA achieves 98.27% accuracy with only 10% training data, significantly outperforming previous state-of-the-art methods. This demonstrates EVA’s transformative potential in enabling efficient and scalable medical image classification.

Once again, we would like to extend our heartfelt congratulations to all nominees and recipient of the Best Paper Award for their outstanding contributions.

Read the full paper here.
More details on the ACM MM 2024 Best Paper Awards.