Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2603.29768 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911557996249088 |
|---|---|
| author | Liao, Jing-Xiao Wang, Haoran Li, Tao Lyu, Daoming Zhang, Yi Cai, Chengjun Fan, Feng-Lei |
| author_facet | Liao, Jing-Xiao Wang, Haoran Li, Tao Lyu, Daoming Zhang, Yi Cai, Chengjun Fan, Feng-Lei |
| contents | With the development of foundational models, model compression has become a critical requirement. Various model compression approaches have been proposed such as low-rank decomposition, pruning, quantization, ergodic dynamic systems, and knowledge distillation, which are based on different heuristics. To elevate the field from fragmentation to a principled discipline, we construct a unifying mathematical framework for model compression grounded in measure theory. We further demonstrate that each model compression technique is mathematically equivalent to a neural network subject to a regularization. Building upon this mathematical and structural equivalence, we propose an experimentally-verified data-free model compression framework, termed \textit{Big2Small}, which translates Implicit Neural Representations (INRs) from data domain to the domain of network parameters. \textit{Big2Small} trains compact INRs to encode the weights of larger models and reconstruct the weights during inference. To enhance reconstruction fidelity, we introduce Outlier-Aware Preprocessing to handle extreme weight values and a Frequency-Aware Loss function to preserve high-frequency details. Experiments on image classification and segmentation demonstrate that \textit{Big2Small} achieves competitive accuracy and compression ratios compared to state-of-the-art baselines. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_29768 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Big2Small: A Unifying Neural Network Framework for Model Compression Liao, Jing-Xiao Wang, Haoran Li, Tao Lyu, Daoming Zhang, Yi Cai, Chengjun Fan, Feng-Lei Machine Learning With the development of foundational models, model compression has become a critical requirement. Various model compression approaches have been proposed such as low-rank decomposition, pruning, quantization, ergodic dynamic systems, and knowledge distillation, which are based on different heuristics. To elevate the field from fragmentation to a principled discipline, we construct a unifying mathematical framework for model compression grounded in measure theory. We further demonstrate that each model compression technique is mathematically equivalent to a neural network subject to a regularization. Building upon this mathematical and structural equivalence, we propose an experimentally-verified data-free model compression framework, termed \textit{Big2Small}, which translates Implicit Neural Representations (INRs) from data domain to the domain of network parameters. \textit{Big2Small} trains compact INRs to encode the weights of larger models and reconstruct the weights during inference. To enhance reconstruction fidelity, we introduce Outlier-Aware Preprocessing to handle extreme weight values and a Frequency-Aware Loss function to preserve high-frequency details. Experiments on image classification and segmentation demonstrate that \textit{Big2Small} achieves competitive accuracy and compression ratios compared to state-of-the-art baselines. |
| title | Big2Small: A Unifying Neural Network Framework for Model Compression |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2603.29768 |