The advancement of large language models (LLMs) and multi-modal LLMs (MLLMs) has historically relied on scaling model parameters. However, as hardware limits constrain further model growth, the primary computational bottleneck has shifted to the quadratic cost of self-attention over increasingly long sequences by ultra-long text contexts, high-resolution images, and extended videos. In this position paper, \textbf{we argue that the focus of research for efficient artificial intelligence (AI) is shifting from model-centric compression to data-centric compression}. We position data-centric compression as the emerging paradigm, which improves AI efficiency by directly compressing the volume of data processed during model training or inference. To formalize this shift, we establish a unified framework for existing efficiency strategies and demonstrate why it constitutes a crucial paradigm change for long-context AI. We then systematically review the landscape of data-centric compression methods, analyzing their benefits across diverse scenarios. Finally, we outline key challenges and promising future research directions. Our work aims to provide a novel perspective on AI efficiency, synthesize existing efforts, and catalyze innovation to address the challenges posed by ever-increasing context lengths.
@article{arxiv.2505.19147,
title = {Shifting AI Efficiency From Model-Centric to Data-Centric Compression},
author = {Xuyang Liu and Zichen Wen and Shaobo Wang and Junjie Chen and Zhishan Tao and Yubo Wang and Tailai Chen and Xiangqi Jin and Chang Zou and Yiyu Wang and Chenfei Liao and Xu Zheng and Honggang Chen and Weijia Li and Xuming Hu and Conghui He and Linfeng Zhang},
journal= {arXiv preprint arXiv:2505.19147},
year = {2025}
}