Data-Efficient Stream-Based Active Distillation for Scalable Edge Model Deployment
Not on the map. Why?
Abstract (arXiv)
Edge camera-based systems are continuously expanding, facing ever-evolving environments that require regular model updates. In practice, complex teacher models are run on a central server to annotate data, which is then used to train smaller models tailored to the edge devices with limited computational power. This work explores how to select the most useful images for training to maximize model quality while keeping transmission costs low. Our work shows that, for a similar training load (i.e., iterations), a high-confidence stream-based strategy coupled with a diversity-based approach produces a high-quality model with minimal dataset queries.
Key facts. No citing papers in the map yet (data through 13 Oct 2026). Builds on 14 of its 29 references that are in the map. First posted on arXiv 24 Sep 2025.
Builds on
14 of 29 references.
References · 14
- 2014Microsoft COCO: Common Objects in Contextcited by 17,550
- 2015Distilling the Knowledge in a Neural Networkcited by 10,180
- 2019Pseudo-Labeling and Confirmation Bias in Deep Semi-Supervised Learningcited by 509
- 2023A Comprehensive Survey on Test-Time Adaptation under Distribution Shiftscited by 355
- 2019Selection via Proxy: Efficient Data Selection for Deep Learningcited by 339
- 2020Identifying Mislabeled Data using the Area Under the Margin Rankingcited by 260
- 2018Turning Big data into tiny data: Constant-size coresets for k-means, PCA and projective clusteringcited by 94
- 2017Deep Active Learning over the Long Tailcited by 82
- 2022Entropy-based Active Learning for Object Detection with Progressive Diversity Constraintcited by 53
- 2022Plug and Play Active Learning for Object Detectioncited by 28
- 2024Coreset Selection for Object Detectioncited by 18
- 2025Training-Free Dataset Pruning for Instance Segmentationcited by 7
Built on by
Not cited yet.