Open access
Jul 2026
Deep Learning Benchmarks for short-term Arctic Sea Ice Forecasting
This study benchmarks CNN backbone and Transformer backbone families for 10-day sea ice forecasting and indicates that CNN-based models, including PoolFormer, generally show more favorable performance in boundary preservation and prediction stability than Transformer-based models.
Bo-Ram Kim, Junhwa Chi
· The International Archives o... · 0 citations