Bilevel ZOFO: Bridging Parameter-Efficient and Zeroth-Order Techniques for Efficient LLM Fine-Tuning and Meta-Training

• URL: http://arxiv.org/abs/2502.03604v2
• Date: Mon, 27 Oct 2025 02:10:58 GMT
• Title: Bilevel ZOFO: Bridging Parameter-Efficient and Zeroth-Order Techniques for Efficient LLM Fine-Tuning and Meta-Training
• Authors: Reza Shirkavand, Peiran Yu, Qi He, Heng Huang,
• Abstract summary: Fine-tuning pre-trained Large Language Models (LLMs) for downstream tasks presents significant computational challenges.<n>We propose Bilevel-ZOFO, a bilevel optimization method that couples fast, local FO-PEFT adaptation at the inner level with stable, memory-efficient ZO updates of the full backbone at the outer level.<n>We show that Bilevel-ZOFO significantly outperforms existing ZO and FO-PEFT methods, achieving 2-4 times faster training while maintaining similar memory efficiency.
• Score: 44.89297451402362
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Fine-tuning pre-trained Large Language Models (LLMs) for downstream tasks using First-Order (FO) optimizers presents significant computational challenges. Parameter-Efficient Fine-Tuning (PEFT) methods address these by freezing most model parameters and training only a small subset. However, PEFT often underperforms compared to full fine-tuning when high task-specific accuracy is required. Zeroth-Order (ZO) methods fine-tune the entire pre-trained model without back-propagation, estimating gradients through forward passes only. While memory-efficient, ZO methods suffer from slow convergence and high sensitivity to prompt selection. We bridge these two worlds with Bilevel-ZOFO, a bilevel optimization method that couples fast, local FO-PEFT adaptation at the inner level with stable, memory-efficient ZO updates of the full backbone at the outer level. The FO-PEFT inner loop performs fast, low-memory local adaptation that reduces the variance of ZO estimates and stabilizes the search, guiding the outer ZO updates of the full backbone and reducing prompt sensitivity. In the mean time, the outer ZO provides better generalization ability for PEFT. We provide theoretical convergence guarantees and empirically demonstrate that Bilevel-ZOFO significantly outperforms existing ZO and FO-PEFT methods, achieving 2-4 times faster training while maintaining similar memory efficiency. Additionally, we show by updating the backbone with ZO and adapting only a tiny FO-PEFT block per task, Bilevel-ZOFO combines full-model capacity with few-shot efficiency, making it a very efficient meta-learning algorithm that quickly adapts to new tasks.

Related papers

• Hi-ZFO: Hierarchical Zeroth- and First-Order LLM Fine-Tuning via Importance-Guided Tensor Selection [4.808936079900314]
  We propose textbfHi-ZFO (textbfHierarchical textbfZeroth- and textbfFirst-textbfOrder optimization) to synergize FO gradients with ZO estimation.<n>We show that Hi-ZFO consistently achieves superior performance while significantly reducing the training time.
  arXiv  Detail & Related papers  (2026-01-09T03:20:54Z)
• FRoD: Full-Rank Efficient Fine-Tuning with Rotational Degrees for Fast Convergence [20.138989330054955]
  We propose FRoD, a novel fine-tuning method that combines hierarchical joint decomposition with rotational degrees of freedom.<n>On 20 benchmarks spanning vision, reasoning, and language understanding, FRoD matches full model fine-tuning in accuracy, while using only 1.72% of trainable parameters under identical training budgets.
  arXiv  Detail & Related papers  (2025-12-29T14:13:45Z)
• Optimization-Inspired Few-Shot Adaptation for Large Language Models [25.439708260502556]
  Large Language Models (LLMs) have demonstrated remarkable performance in real-world applications.<n>Adapting LLMs to novel tasks via fine-tuning often requires substantial training data and computational resources that are impractical in few-shot scenarios.<n>Existing approaches, such as in-context learning and.<n>Efficient Fine-Tuning (PEFT), face key limitations.
  arXiv  Detail & Related papers  (2025-05-25T11:54:23Z)
• Training Deep Learning Models with Norm-Constrained LMOs [56.00317694850397]
  We study optimization methods that leverage the linear minimization oracle (LMO) over a norm-ball. We propose a new family of algorithms that uses the LMO to adapt to the geometry of the problem and, perhaps surprisingly, show that they can be applied to unconstrained problems.
  arXiv  Detail & Related papers  (2025-02-11T13:10:34Z)
• Harmony in Divergence: Towards Fast, Accurate, and Memory-efficient Zeroth-order LLM Fine-tuning [44.907586955452295]
  Large language models (LLMs) excel across various tasks, but standard first-order (FO) fine-tuning demands considerable memory.<n>Recently, zeroth-order (ZO) optimization stood out as a promising memory-efficient training paradigm.<n>We introduce a novel layer-wise divergence analysis that uncovers the distinct update pattern of FO and ZO optimization.<n>Our results demonstrate that DiZO significantly reduces the needed iterations for convergence without sacrificing throughput.
  arXiv  Detail & Related papers  (2025-02-05T16:03:17Z)
• Refining Salience-Aware Sparse Fine-Tuning Strategies for Language Models [14.68920095399595]
  sparsity-based PEFT (SPEFT) introduces trainable sparse adaptations to the weight matrices in the model.<n>We conduct the first systematic evaluation of salience metrics for SPEFT, inspired by zero-cost NAS proxies.<n>Our work challenges the notion that complexity is necessary for effective PEFT.
  arXiv  Detail & Related papers  (2024-12-18T04:14:35Z)
• ALoRE: Efficient Visual Adaptation via Aggregating Low Rank Experts [71.91042186338163]
  ALoRE is a novel PETL method that reuses the hypercomplex parameterized space constructed by Kronecker product to Aggregate Low Rank Experts.<n>Thanks to the artful design, ALoRE maintains negligible extra parameters and can be effortlessly merged into the frozen backbone.
  arXiv  Detail & Related papers  (2024-12-11T12:31:30Z)
• Layer-wise Importance Matters: Less Memory for Better Performance in Parameter-efficient Fine-tuning of Large Language Models [19.163639128631534]
  Importance-aware Sparse Tuning (IST) is a plug-and-play technique compatible with various PEFT methods that operate on a per-layer basis. IST dynamically updates selected layers in PEFT modules, leading to reduced memory demands.
  arXiv  Detail & Related papers  (2024-10-15T16:53:26Z)
• Simultaneous Computation and Memory Efficient Zeroth-Order Optimizer for Fine-Tuning Large Language Models [33.911521719528686]
  Fine-tuning is powerful for adapting large language models to downstream tasks, but it often results in huge memory usages. A promising approach is using Zeroth-Order (ZO) gradients, which estimates to replace First-Order (FO) gradients. We introduce a novel layer-wise sparse computation and memory efficient ZO, named LeZO.
  arXiv  Detail & Related papers  (2024-10-13T12:47:37Z)
• Enhancing Zeroth-order Fine-tuning for Language Models with Low-rank Structures [21.18741772731095]
  Zeroth-order (ZO) algorithms offer a promising alternative by approximating gradients using finite differences of function values. Existing ZO methods struggle to capture the low-rank gradient structure common in LLM fine-tuning, leading to suboptimal performance. This paper proposes a low-rank ZO algorithm (LOZO) that effectively captures this structure in LLMs.
  arXiv  Detail & Related papers  (2024-10-10T08:10:53Z)
• BIPEFT: Budget-Guided Iterative Search for Parameter Efficient Fine-Tuning of Large Pretrained Language Models [63.52035708182815]
  We introduce a novel Budget-guided Iterative search strategy for automatic PEFT (BIPEFT) BIPEFT employs a new iterative search strategy to disentangle the binary module and rank dimension search spaces. Extensive experiments on public benchmarks demonstrate the superior performance of BIPEFT for downstream tasks with a low parameter budget.
  arXiv  Detail & Related papers  (2024-10-04T18:50:46Z)
• SHERL: Synthesizing High Accuracy and Efficient Memory for Resource-Limited Transfer Learning [63.93193829913252]
  We propose an innovative METL strategy called SHERL for resource-limited scenarios. In the early route, intermediate outputs are consolidated via an anti-redundancy operation. In the late route, utilizing minimal late pre-trained layers could alleviate the peak demand on memory overhead.
  arXiv  Detail & Related papers  (2024-07-10T10:22:35Z)
• Bypass Back-propagation: Optimization-based Structural Pruning for Large Language Models via Policy Gradient [57.9629676017527]
  We propose an optimization-based structural pruning on Large-Language Models. We learn the pruning masks in a probabilistic space directly by optimizing the loss of the pruned model. Our method operates for 2.7 hours with around 35GB memory for the 13B models on a single A100 GPU.
  arXiv  Detail & Related papers  (2024-06-15T09:31:03Z)
• Revisiting Zeroth-Order Optimization for Memory-Efficient LLM Fine-Tuning: A Benchmark [166.40879020706151]
  This paper proposes a shift towards BP-free, zeroth-order (ZO) optimization as a solution for reducing memory costs during fine-tuning. Unlike traditional ZO-SGD methods, our work expands the exploration to a wider array of ZO optimization techniques. Our study unveils previously overlooked optimization principles, highlighting the importance of task alignment, the role of the forward gradient method, and the balance between algorithm complexity and fine-tuning performance.
  arXiv  Detail & Related papers  (2024-02-18T14:08:48Z)
• Federated Learning of Large Language Models with Parameter-Efficient Prompt Tuning and Adaptive Optimization [71.87335804334616]
  Federated learning (FL) is a promising paradigm to enable collaborative model training with decentralized data. The training process of Large Language Models (LLMs) generally incurs the update of significant parameters. This paper proposes an efficient partial prompt tuning approach to improve performance and efficiency simultaneously.
  arXiv  Detail & Related papers  (2023-10-23T16:37:59Z)
• Federated Hypergradient Descent [0.0]
  We apply a principled approach on a method for adaptive client learning rate, number of local steps, and batch size. In our federated learning applications, our primary motivations are minimizing communication budget as well as local computational resources in the training pipeline. We show our numerical results through extensive empirical experiments with the Federated EMNIST-62 (FEMNIST) and Federated Stack Overflow (FSO) datasets.
  arXiv  Detail & Related papers  (2022-11-03T19:22:00Z)
• BOME! Bilevel Optimization Made Easy: A Simple First-Order Approach [46.457298683984924]
  Bilevel optimization (BO) is useful for solving a variety important machine learning problems. Conventional methods need to differentiate through the low-level optimization process with implicit differentiation. First-order BO depends only on first-order information, requires no implicit differentiation.
  arXiv  Detail & Related papers  (2022-09-19T01:51:12Z)
• Few-Shot Parameter-Efficient Fine-Tuning is Better and Cheaper than In-Context Learning [81.3514358542452]
  Few-shot in-context learning (ICL) incurs substantial computational, memory, and storage costs because it involves processing all of the training examples every time a prediction is made. parameter-efficient fine-tuning offers an alternative paradigm where a small set of parameters are trained to enable a model to perform the new task. In this paper, we rigorously compare few-shot ICL and parameter-efficient fine-tuning and demonstrate that the latter offers better accuracy as well as dramatically lower computational costs.
  arXiv  Detail & Related papers  (2022-05-11T17:10:41Z)
• Model-Agnostic Multitask Fine-tuning for Few-shot Vision-Language Transfer Learning [59.38343286807997]
  We propose Model-Agnostic Multitask Fine-tuning (MAMF) for vision-language models on unseen tasks. Compared with model-agnostic meta-learning (MAML), MAMF discards the bi-level optimization and uses only first-order gradients. We show that MAMF consistently outperforms the classical fine-tuning method for few-shot transfer learning on five benchmark datasets.
  arXiv  Detail & Related papers  (2022-03-09T17:26:53Z)

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.