Summary of A Survey on Efficient Inference For Large Language Models, by Zixuan Zhou et al.
A Survey on Efficient Inference for Large Language Models
by Zixuan Zhou, Xuefei Ning, Ke Hong, Tianyu Fu, Jiaming Xu, Shiyao Li, Yuming Lou, Luning Wang, Zhihang Yuan, Xiuhong Li, Shengen Yan, Guohao Dai, Xiao-Ping Zhang, Yuhan Dong, Yu Wang
First submitted to arxiv on: 22 Apr 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This comprehensive survey aims to enhance the efficiency of Large Language Model (LLM) inference by analyzing the primary causes of inefficient LLM inference, including large model size, quadratic-complexity attention operation, and auto-regressive decoding approach. The paper presents a taxonomy that organizes existing literature into data-level, model-level, and system-level optimization techniques. Comparative experiments on representative methods within critical sub-fields provide quantitative insights. The study concludes with knowledge summary and future research directions. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This survey helps us understand why Large Language Models are slow to use and how we can make them faster. It looks at the main reasons for this slowness, like big models, complex calculations, and how language is generated. The paper organizes current ideas into three categories: changing the data, improving the model, and optimizing the system. By comparing different methods, researchers can see what works best in certain areas. Finally, the study concludes with a summary of what we know now and where we might go from here. |
Keywords
» Artificial intelligence » Attention » Inference » Large language model » Optimization
