English

Fly-CL: A Fly-Inspired Framework for Enhancing Efficient Decorrelation and Reduced Training Time in Pre-trained Model-based Continual Representation Learning

Machine Learning 2026-03-03 v2 Artificial Intelligence Computer Vision and Pattern Recognition

Abstract

Using a nearly-frozen pretrained model, the continual representation learning paradigm reframes parameter updates as a similarity-matching problem to mitigate catastrophic forgetting. However, directly leveraging pretrained features for downstream tasks often suffers from multicollinearity in the similarity-matching stage, and more advanced methods can be computationally prohibitive for real-time, low-latency applications. Inspired by the fly olfactory circuit, we propose Fly-CL, a bio-inspired framework compatible with a wide range of pretrained backbones. Fly-CL substantially reduces training time while achieving performance comparable to or exceeding that of current state-of-the-art methods. We theoretically show how Fly-CL progressively resolves multicollinearity, enabling more effective similarity matching with low time complexity. Extensive simulation experiments across diverse network architectures and data regimes validate Fly-CL's effectiveness in addressing this challenge through a biologically inspired design. Code is available at https://github.com/gfyddha/Fly-CL.

Keywords

Cite

@article{arxiv.2510.16877,
  title  = {Fly-CL: A Fly-Inspired Framework for Enhancing Efficient Decorrelation and Reduced Training Time in Pre-trained Model-based Continual Representation Learning},
  author = {Heming Zou and Yunliang Zang and Wutong Xu and Xiangyang Ji},
  journal= {arXiv preprint arXiv:2510.16877},
  year   = {2026}
}

Comments

ICLR 2026 accepted paper

R2 v1 2026-07-01T06:45:47.997Z