@inproceedings{siddiqui-etal-2025-llms,
    title = "{LLM}s on a Budget? Say {HOLA}",
    author = "Siddiqui, Zohaib Hasan  and
      Gao, Jiechao  and
      Shabbir, Ebad  and
      Azeez, Mohammad Anas  and
      Ali, Rafiq  and
      Kashyap, Gautam Siddharth  and
      Naseem, Usman",
    editor = "Potdar, Saloni  and
      Rojas-Barahona, Lina  and
      Montella, Sebastien",
    booktitle = "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing: Industry Track",
    month = nov,
    year = "2025",
    address = "Suzhou (China)",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-emnlp/2025.emnlp-industry.71/",
    pages = "1035--1043",
    ISBN = "979-8-89176-333-3",
    abstract = "Running Large Language Models (LLMs) on edge devices is constrained by high compute and memory demands{---}posing a barrier for real-time applications in industries like healthcare, education, and embedded systems. Current solutions such as quantization, pruning, and Retrieval-Augmented Generation (RAG) offer only partial optimizations and often compromise on speed or accuracy. We introduce HOLA, an end-to-end optimization framework for efficient LLM deployment. Internally, it leverages Hierarchical Speculative Decoding (HSD) for faster inference without quality loss. Externally, AdaComp-RAG adjusts retrieval complexity based on context needs. Together with Lo-Bi, which blends structured pruning (LoRA) and quantization, HOLA delivers significant gains: +17.6{\%} EMA on GSM8K, +10.5{\%} MCA on ARC, and reduced latency and memory on edge devices like Jetson Nano{---}proving both scalable and production-ready. Our code is available at: https://github.com/zohaibhasan066/HOLA{\_}Codebase"
}