AI RESEARCH
AHC: Meta-Learned Adaptive Compression for Continual Object Detection on Memory-Constrained Microcontrollers
arXiv CS.AI
•
ArXi:2604.09576v1 Announce Type: new Deploying continual object detection on microcontrollers (MCUs) with under 100KB memory requires efficient feature compression that can adapt to evolving task distributions. Existing approaches rely on fixed compression strategies (e.g., FiLM conditioning) that cannot adapt to heterogeneous task characteristics, leading to suboptimal memory utilization and catastrophic forgetting. We