@inproceedings{ning-etal-2025-efficient,
    title = "Efficient Inference for Large Language Models {--}Algorithm, Model, and System",
    author = "Ning, Xuefei  and
      Dai, Guohao  and
      Bai, Haoli  and
      Hou, Lu  and
      Wang, Yu",
    editor = "Pyatkin, Valentina  and
      Vlachos, Andreas",
    booktitle = "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing: Tutorial Abstracts",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-emnlp/2025.emnlp-tutorials.1/",
    pages = "1--3",
    ISBN = "979-8-89176-336-4",
    abstract = "The inference of LLMs incurs high computational costs, memory access overhead, and memory usage, leading to inefficiencies in terms of latency, throughput, power consumption, and storage. To this end, this tutorial focuses on the increasingly important topic of Efficient Inference for LLMs and aims to provide a systematic understanding of key facts and methodologies from a designer{'}s perspective. We start by introducing the basic concepts of modern LLMs, software and hardware. Following this, we define the efficiency optimization problem. To equip the audience with a designer{'}s mindset, we briefly explain how to diagnose efficiency bottlenecks for a given workload on specific hardware. After introducing the basics, we will introduce our full-stack taxonomy of efficient inference methods for LLMs. We will walk through each category of methodology, using one to three representative methods as examples for each leaf subcategory, elaborating on the design logic behind each method and which inefficiency factors they primarily address. Finally, we will wrap up with a takeaway summary, and future research directions."
}