mlsanityData to 14 Oct 2026

MECAIL: Communication-Aware Incremental Learning for Object Detection with 14.6 KB Spatiotemporal Experts

arXiv 2609.24455 ↗First posted 21 Sep 2026cs.CV cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Intelligent transportation systems require Incremental Learning (IL) to continually improve their overall performance in dynamic environments. However, most edge devices lack the computational resources to support on-device IL, requiring updates to be transmitted from centralized servers. We propose using this setup to obtain dense, specialized module coverage that adapts a fixed base model to specific spatiotemporal contexts, such as parking lots, gas stations, ferries, or construction sites. However, in order to reliably transmit these modules to the edge device, using TCP, UDP, and BTP over V2X, Wi-Fi, and 2G-5G hardware, we establish a strict limit of 14.6 KB per module to fit within the first TCP window and to minimize UDP/BTP fragmentation. We further introduce Mixture-of-Experts for Communication-Aware Incremental Learning (MECAIL), the first method that meets this strict requirement, in which each new domain or environment is served by a small expert network that adapts the base model. We validate MECAIL on D-RICO and ODinW-13, where it largely matches the performance of parameter-heavy approaches while enabling practical, bandwidth-efficient large-scale deployment. This allows comprehensive coverage by experts for highly specific, focused, and temporary situations.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 35 of its 53 references that are in the map. First posted on arXiv 21 Sep 2026.

Builds on

35 of 53 references.

References · 35

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  2. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  3. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,139
  4. 2016Overcoming catastrophic forgetting in neural networkscited by 4,787
  5. 2019Parameter-Efficient Transfer Learning for NLPcited by 4,236
  6. 2015Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classificationcited by 3,611
  7. 2023Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detectioncited by 3,332
  8. 2017Gradient Episodic Memory for Continual Learningcited by 1,898
  9. 2023A Comprehensive Survey of Continual Learning: Theory, Method and Applicationcited by 872
  10. 2013Network In Networkcited by 605
  11. 2016Expert Gate: Lifelong Learning with a Network of Expertscited by 372
  12. 2020Frustratingly Simple Few-Shot Object Detectioncited by 282
23 more
  1. 2024Boosting Continual Learning of Vision-Language Models via Mixture-of-Experts Adapterscited by 199
  2. 2022ELEVATER: A Benchmark and Toolkit for Evaluating Language-Augmented Visual Modelscited by 141
  3. 2020A Neural Dirichlet Process Mixture Model for Task-Free Continual Learningcited by 134
  4. 2020Incremental Object Detection via Meta-Learningcited by 58
  5. 2024Theory on Mixture-of-Experts in Continual Learningcited by 54
  6. 2024Mixture of Experts Meets Prompt-Based Continual Learningcited by 43
  7. 2022Continual Object Detection: A review of definitions, strategies, and challengescited by 25
  8. 2023FedET: A Communication-Efficient Federated Class-Incremental Learning Framework Based on Enhanced Transformercited by 23
  9. 2023Three ways to improve feature alignment for open vocabulary detectioncited by 17
  10. 2024Bridge Past and Future: Overcoming Information Asymmetry in Incremental Object Detectioncited by 16
  11. 2024Zero-shot Generalizable Incremental Learning for Vision-Language Object Detectioncited by 15
  12. 2022Incremental-DETR: Incremental Few-Shot Object Detection via Self-Supervised Learningcited by 14
  13. 2025Federated Continual Learning: Concepts, Challenges, and Solutionscited by 13
  14. 2024Learning More Generalized Experts by Merging Experts in Mixture-of-Expertscited by 12
  15. 2021Lifelong Infinite Mixture Model Based on Knowledge-Driven Dirichlet Processcited by 7
  16. 2024Replay Consolidation with Label Propagation for Continual Object Detectioncited by 6
  17. 2025RICO: Two Realistic Benchmarks and an In-Depth Analysis for Incremental Learning in Object Detectioncited by 5
  18. 2025DitHub: A Modular Framework for Incremental Open-Vocabulary Object Detectioncited by 3
  19. 2025Textual Inversion for Efficient Adaptation of Open-Vocabulary Object Detectors Without Forgettingcited by 3
  20. 2023CoDeC: Communication-Efficient Decentralized Continual Learningcited by 2
  21. 2025Incremental Object Detection with Prompt-based Methodscited by 2
  22. 2020Empirical Models for the Realistic Generation of Cooperative Awareness Messages in Vehicular Networkscited by 1
  23. 2024Approximate Wireless Communication for Lossy Gradient Updates in IoT Federated Learningcited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2024Dynamic Object Queries for Transformer-based Incremental Object Detection
  2. 2025Video Object Recognition in Mobile Edge Networks: Local Tracking or Edge Detection?
  3. 2024CASA: Class-Agnostic Shared Attributes in Vision-Language Models for Efficient Incremental Object Detection
  4. 2025Exploiting Stability-Plasticity Asymmetry in Pretrained Detectors for Incremental Object Detection
  5. 2023Leveraging the Edge and Cloud for V2X-Based Real-Time Object Detection in Autonomous Driving
  6. 2025EffiComm: Bandwidth Efficient Multi Agent Communication
  7. 2026Collision-Aware Vision-Language Learning for End-to-End Driving with Multimodal Infraction Datasets
  8. 2026Edge AI for Automotive Vulnerable Road User Safety: Deployable Detection via Knowledge Distillation