Aligning Large Language Models With Human Feedback: Mathematical foundations and algorithm design [Special Issue on the Mathematics of Deep Learning]

Siliang Zeng, Luca Viano, Chenliang Li, Jiaxiang Li, Markus Wulfmeier, Stefano Ermon, Alfredo García, Mingyi Hong

Abstract

This article provides an introduction to the mathematical foundations and algorithmic frameworks used to align Large Language Models (LLMs) with human intentions, preferences, and values. We discuss standard alignment techniques, such as fine-tuning (SFT), reinforcement learning with human feedback (RLHF), and direct preference optimization (DPO). We also explore the theoretical underpinnings of learning from human preferences, drawing connections to inverse reinforcement learning (IRL) and discrete choice models. We present state-of-the-art algorithms in a tutorial style, discuss their advantages and limitations, and offer insights into practical implementation. Our exposition is intended to serve as a comprehensive resource for researchers and practitioners, providing both a foundational understanding of alignment methodologies and a framework for developing more robust and scalable alignment techniques.
BibTeX
@article{zeng2026aligning,
  author = {Siliang Zeng and Luca Viano and Chenliang Li and Jiaxiang Li and Markus Wulfmeier and Stefano Ermon and Alfredo García and Mingyi Hong},
  title = {{Aligning Large Language Models With Human Feedback: Mathematical foundations and algorithm design [Special Issue on the Mathematics of Deep Learning]}},
  journal = {IEEE Signal Process. Mag.},
  year = {2026},
  doi = {10.1109/msp.2026.3666824},
}

Loading related content...

Figure