Metraipolis › archive › dc_32ce79e239e0
APQF: Agentic Profiling-Guided Structured Pruning and Mixed-Precision Quantization with Adaptive Fine-Tuning — arXiv:2608.05499
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.05499
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-31T10:35:57.182Z
- SHA-256 of the archived text
5a6071c00d5c6af0c2c2008b686901855bd8ff43f8749b1ee851150cdd8e8d82- Stable URL of this page
https://metraipolis.com/documento/dc_32ce79e239e0
How to cite
APQF: Agentic Profiling-Guided Structured Pruning and Mixed-Precision Quantization with Adaptive Fine-Tuning — arXiv:2608.05499 — arXiv (en), https://arxiv.org/abs/2608.05499 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_32ce79e239e0 (Metraipolis archive dc_32ce79e239e0, sha256 5a6071c00d5c6af0c2c2008b686901855bd8ff43f8749b1ee851150cdd8e8d82)
BibTeX
@misc{dc32ce79e239e0,
title = {APQF: Agentic Profiling-Guided Structured Pruning and Mixed-Precision Quantization with Adaptive Fine-Tuning — arXiv:2608.05499},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.05499},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_32ce79e239e0; sha256 5a6071c00d5c6af0c2c2008b686901855bd8ff43f8749b1ee851150cdd8e8d82},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.05499. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Modern deep neural networks achieve strong performance, but their scale makes them costly and slow, especially on resource-constrained edge devices. Pruning and quantization address this, but rely on manual, expert choices and on algorithms that are hard to apply across architectures. Uniform settings also ignore how differently individual layers respond to compression, which costs accuracy. We introduce APQF, an agentic profiling-guided framework that combines structured pruning, mixed-precision quantization-aware training, and accuracy recovery in one automated pipeline. A profiling agent measures how cost is distributed across the model and how sensitive each part is to pruning, and this evidence drives per-layer pruning ratios, per-layer bit-widths, and the recovery strategy, all proposed by LLM planners and validated before execution. To our knowledge, APQF is the first framework to combine LLM-guided, profiling-grounded decisions with a fully training-aware pruning and quantization pipeline for both CNNs and vision transformers. We evaluate APQF on ResNet, VGG7, ViT, DeiT, and Swin using ImageNet-1k and CIFAR-10. On ImageNet it cuts compute to 5.6-7.7 percent of the original bit-operations, a 13-18x reduction, while keeping accuracy close to the baseline, and under a 200K-image budget it stays roughly 17 points higher in Top-1 than existing joint pruning and quantization methods. On CIFAR-10 it compresses further than that method on four of five architectures. On VGG7 it reaches 93.15 percent using only 0.41 percent of baseline bit-operations, the only method at that compression level to improve on its full-precision baseline. Ablations show that uniform compression loses the most accuracy at matched compute, and that withholding profiling data from the planner hurts every model. Six LLM planners, including free open-weight ones, all reach 97.4-97.9 percent on Swin-Tiny. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-07 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI); Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2608.05499 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-31T10:23:30.976Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-08-07?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).