LGOCJun 4, 2025

Leveraging Coordinate Momentum in SignSGD and Muon: Memory-Optimized Zero-Order

arXiv:2506.04430v47 citationsh-index: 18Has Code
Originality Incremental advance
AI Analysis

This work addresses resource constraints in LLM adaptation, offering a practical solution for fine-tuning with reduced memory overhead, though it is incremental as it builds on existing zero-order and momentum-based techniques.

The paper tackles the high memory and computational costs of fine-tuning large language models (LLMs) by proposing zero-order optimization methods, specifically JAGUAR SignSGD and JAGUAR Muon, which achieve convergence comparable to first-order methods while significantly reducing memory usage.

Fine-tuning Large Language Models (LLMs) is essential for adapting pre-trained models to downstream tasks. Yet traditional first-order optimizers such as Stochastic Gradient Descent (SGD) and Adam incur prohibitive memory and computational costs that scale poorly with model size. In this paper, we investigate zero-order (ZO) optimization methods as a memory- and compute-efficient alternative, particularly in the context of parameter-efficient fine-tuning techniques like LoRA. We propose $\texttt{JAGUAR SignSGD}$, a ZO momentum-based algorithm that extends ZO SignSGD, requiring the same number of parameters as the standard ZO SGD and only $\mathcal{O}(1)$ function evaluations per iteration. To the best of our knowledge, this is the first study to establish rigorous convergence guarantees for SignSGD in the stochastic ZO case. We further propose $\texttt{JAGUAR Muon}$, a novel ZO extension of the Muon optimizer that leverages the matrix structure of model parameters, and we provide its convergence rate under arbitrary stochastic noise. Through extensive experiments on challenging LLM fine-tuning benchmarks, we demonstrate that the proposed algorithms meet or exceed the convergence quality of standard first-order methods, achieving significant memory reduction. Our theoretical and empirical results establish new ZO optimization methods as a practical and theoretically grounded approach for resource-constrained LLM adaptation. Our code is available at https://github.com/brain-mmo-lab/ZO_LLM

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes