Enhancement Optimization Techniques on Large Language Model

Dear Colleagues,

The rapid evolution of Large Language Models (LLMs) has historically been dominated by a paradigm of scaling, i.e., increasing model size and data volume to unlock new capabilities. However, this pursuit of sheer scale has revealed significant practical limitations, including exorbitant computational costs, challenges in deployment, and ongoing concerns regarding reliability and safety. While the existing literature often celebrates scaling laws and novel architectures, a critical next frontier has emerged: moving beyond scale to achieve superior performance through strategic refinement. It is against this backdrop that this Special Issue is situated. It will focus on interdisciplinary research that performs the essential work of making LLMs more efficient, capable, and deployable.

The purpose of this Special Issue is to consolidate and advance the field by providing a dedicated venue for work that bridges the gap between theoretical potential and practical application. By curating state-of-the-art solutions in algorithmic efficiency, reasoning, safety, and data-centric optimization, we aim to directly complement initial model development and serve as a foundational resource for enabling the next generation of sustainable, robust, and accessible LLMs.

Relevant topics for this Special Issue include, but are not limited to, the following areas:

• Model compression and pruning;
• Quantization;
• Hardware-aware optimization;
• Data synthesis and augmentation;
• Efficient inference serving;
• Caching and speculative decoding;
• Resource management for training and inference;
• Edge and on-device deployment;
• Bias mitigation and fairness;
• Interpretability and Explainability (XAI);
• Safety alignment frameworks.

Dr. Shanjiang Tang
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Big Data and Cognitive Computing is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1800 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• efficient AI
• model compression reasoning enhancement 
• parameter-efficient fine-tuning (PEFT) 
• AI alignment 
• retrieval-augmented generation (RAG)