SubZero+: Efficient Zeroth-Order LLM Fine-Tuning via Large Learning Rates

• URL: http://arxiv.org/abs/2608.15665v1
• Date: Sun, 16 Aug 2026 10:13:25 GMT
• Title: SubZero+: Efficient Zeroth-Order LLM Fine-Tuning via Large Learning Rates
• Abstract summary: SubZero+ is an improved SubZero framework that improves stability in three complementary ways.<n>We show that SubZero+ consistently outperforms prior ZO baselines, enlarges the stable learning-rate range, and narrows the gap to first-order methods with minimal extra memory overhead.
• Score: 25.09609527559127
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Zeroth-order (ZO) optimization enables backpropagation-free fine-tuning of large language models, but existing ZO methods suffer from high-variance gradient estimators, making convergence unstable and highly sensitive to learning rates. We propose SubZero+, an improved SubZero framework that improves stability in three complementary ways: (i) multi-query gradient estimation within layer-specific low-rank subspaces to reduce variance without exhibiting the multi-query paradox; (ii) a subspace Adam optimizer that performs adaptive updates using in-subspace multi-query gradient statistics; and (iii) a sign correction for QR-based subspace construction to ensure Haar-distributed projection matrices, eliminating implementation-dependent orientation ambiguity. Experiments on models from 1.3B to 32B across SuperGLUE, under both full-parameter tuning and LoRA, show that SubZero+ consistently outperforms prior ZO baselines, enlarges the stable learning-rate range, and narrows the gap to first-order methods with minimal extra memory overhead.

Related papers

• Adaptively Incorporating Directional Hints into Zeroth-Order Optimization [49.940317286021696]
  We develop an algorithm that achieves zeroth-order optimization of non- functions with the aid directional hints.<n>We validate the method, demonstrating sustained progress on non- control tasks.
  arXiv  Detail & Related papers  (2026-09-08T05:42:02Z)
• ZO-Act: Efficient Zeroth-Order Fine-Tuning via One-Shot Activation-Informed Low-Rank Subspaces [8.08601759467653]
  Zeroth-order (ZO) optimization enables fine-tuning large language models when backpropagation is unavailable or memory-prohibitive.<n>We propose ZO-Act, an activation-informed ZO fine-tuning method that restricts perturbations to a fixed low-rank subspace.<n>We analyze ZO-Act as zeroth-order optimization over a restricted coefficient space and show that perturbing the low-dimensional coefficients reduces both the variance-dependent convergence term and the finite-difference error of the ZO estimator.
  arXiv  Detail & Related papers  (2026-07-01T16:12:50Z)
• SDS-LoRA: Overcoming Anisotropic Gradient Scaling in Low-Rank Adaptation [64.33799467286265]
  Low-Rank Adaptation (LoRA) enables efficient adaptation of large pre-trained models to downstream tasks.<n>We show that when a full fine-tuning gradient is backpropagated to the low-rank matrices, it undergoes anisotropic scaling driven by their singular values.<n>We propose a new low-rank parameterization, SDS-LoRA, which structurally decouples singular values from the backward pass.
  arXiv  Detail & Related papers  (2026-06-15T09:27:50Z)
• Denoising-GS: Gaussian Splatting with Spatial-aware Denoising [59.331050242447255]
  3D Gaussian Splatting (3DGS) has achieved remarkable success in high-fidelity Novel View Synthesis (NVS)<n>We introduce a new perspective by formulating the optimization of 3DGS as a primitive denoising process.<n>We propose Denoising-GS, a coherent denoising framework for primitives by taking both the positions and spatial structure into consideration.<n> Experiments conducted on three benchmark datasets demonstrate that Denoising-GS consistently enhances NVS fidelity while maintaining representation compactness.
  arXiv  Detail & Related papers  (2026-05-14T14:24:21Z)
• Powering Up Zeroth-Order Training via Subspace Gradient Orthogonalization [40.95701844244596]
  We show that ZO optimization can be substantially improved by unifying two complementary principles.<n>We instantiate in a new method, ZO-Muon, admitting a natural interpretation as a low-rank Muon in the ZO setting.
  arXiv  Detail & Related papers  (2026-02-19T08:08:33Z)
• Robust and Efficient Zeroth-Order LLM Fine-Tuning via Adaptive Bayesian Subspace Optimizer [4.6561758107970395]
  Fine-tuning large language models (LLMs) with zeroth-order (ZO) optimization reduces memory by approximating gradients through function evaluations.<n>We introduce BSZO, an adaptive textbfBayesian textbfSubspace textbfZeroth-Order textbfOptimizer.
  arXiv  Detail & Related papers  (2026-01-04T09:35:11Z)
• Low-Rank Curvature for Zeroth-Order Optimization in LLM Fine-Tuning [8.349781300731225]
  We introduce LOREN, a curvature-aware zeroth-order (ZO) optimization method for fine-tuning large language models (LLMs)<n>Existing ZO methods, which estimate gradients via finite differences using random perturbations, often suffer from high variance and suboptimal search directions.<n>Our approach addresses these challenges by: (i) adaptively estimating an anisotropic perturbation distribution for gradient estimation, (ii) capturing curvature through a low-rank block diagonal preconditioner, and (iii) applying a REINFORCE leave-one-out (RLOO) gradient estimator to reduce variance.
  arXiv  Detail & Related papers  (2025-11-11T08:34:09Z)
• Regularizing Subspace Redundancy of Low-Rank Adaptation [54.473090597164834]
  We propose ReSoRA, a method that explicitly models redundancy between mapping subspaces and adaptively Regularizes Subspace redundancy of Low-Rank Adaptation.<n>Our proposed method consistently facilitates existing state-of-the-art PETL methods across various backbones and datasets in vision-language retrieval and standard visual classification benchmarks.<n>As a training supervision, ReSoRA can be seamlessly integrated into existing approaches in a plug-and-play manner, with no additional inference costs.
  arXiv  Detail & Related papers  (2025-07-28T11:52:56Z)
• Breaking the Frozen Subspace: Importance Sampling for Low-Rank Optimization in LLM Pretraining [43.39070237124361]
  Low-rank optimization is a promising approach to enabling memory-efficient training of large language models.<n>A key challenge in these methods is selecting suitable subspaces to ensure an effective trajectory.<n>We propose importance sampling for low-rank optimization in LLM pretraining with a provable convergence guarantee.
  arXiv  Detail & Related papers  (2025-02-09T06:30:19Z)
• TeZO: Empowering the Low-Rankness on the Temporal Dimension in the Zeroth-Order Optimization for Fine-tuning LLMs [58.19080159470868]
  We propose a novel low-rank ZO estimator, TeZO, which captures the low-rankness across both the model and temporal dimension.<n>Specifically, we represent ZO perturbations along the temporal dimension as a 3D tensor and employ Canonical Polyadic Decomposition (CPD) to extract each low-rank 2D matrix.
  arXiv  Detail & Related papers  (2025-01-31T11:34:03Z)
• Zeroth-Order Fine-Tuning of LLMs in Random Subspaces [63.10833446782114]
  As language models grow in size, memory demands for backpropagation increase.<n>Zeroth-order (ZO) optimization methods offer a memory-efficient alternative.<n>In this paper, we propose Subspace Zero-order optimization to address the challenges posed by posed by high dimensionality perturbations.
  arXiv  Detail & Related papers  (2024-10-11T17:01:43Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.