GEM: A Scale-Aware and Distribution-Sensitive Sparse Fine-Tuning Framework for Effective Downstream Adaptation

• URL: http://arxiv.org/abs/2508.16191v1
• Date: Fri, 22 Aug 2025 08:12:06 GMT
• Title: GEM: A Scale-Aware and Distribution-Sensitive Sparse Fine-Tuning Framework for Effective Downstream Adaptation
• Abstract summary: Gradient-to-Weight Ratio and Entropy-guided Masking (GEM)<n>GEM prioritizes parameters whose updates are significant in proportion to their initial pre-trained values.<n>It also adaptively determines how many parameters to tune at each layer based on the entropy of parameter values.
• Score: 21.481261295689208
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Parameter-efficient fine-tuning (PEFT) has become a popular way to adapt large pre-trained models to new tasks. Most PEFT methods update only a small subset of parameters while freezing the rest, avoiding redundant computation. As they maximize the absolute size of the updates without regard to the parameters' original scale, the resulting changes in model behavior can be minimal. In contrast, we maximize updates relative to each parameter's scale, yielding more meaningful downstream adaptation. We propose Gradient-to-Weight Ratio and Entropy-guided Masking (GEM), a parameter scale-aware, distribution-sensitive sparse fine-tuning framework. GEM prioritizes parameters whose updates are significant in proportion to their initial pre-trained values. It also adaptively determines how many parameters to tune at each layer based on the entropy of parameter values, thereby making the most effective use of the computational budget in PEFT. Our empirical study demonstrates the efficacy of GEM on both general-domain tasks (GLUE and SuperGLUE) and domain-specific tasks (GSM8k and MBPP), achieving up to a 1.6% improvement in fine-tuning accuracy over full fine-tuning while updating only 0.1% of model parameters.

Related papers

• GRASP: GRouped Activation Shared Parameterization for Parameter-Efficient Fine-Tuning and Robust Inference of Transformers [12.475144734899674]
  We introduce GRASP, a lightweight PEFT framework that partitions the D-dimensional token representations of selected layers into K D groups and learns a shared scaling and shifting vector for each group.<n>We propose StochGRASP, which learns Gaussian distributions as perturbations to the pre-trained weights rather than deterministic values.<n>Under varying levels of noise, StochGRASP consistently outperforms deterministic variants, demonstrating its suitability for energy-efficient and noise-prone hardware platforms.
  arXiv  Detail & Related papers  (2025-12-03T22:17:05Z)
• GaLLoP: Gradient-based Sparse Learning on Low-Magnitude Parameters [20.34415141254838]
  GaLLoP: Gradient-based Sparse Learning on Low-Magnitude Parameters.<n>We introduce a novel sparse fine-tuning technique named GaLLoP: Gradient-based Sparse Learning on Low-Magnitude Parameters.
  arXiv  Detail & Related papers  (2025-10-22T17:11:49Z)
• Gradient-based Fine-Tuning through Pre-trained Model Regularization [20.823624386591902]
  We propose an efficient gradient-based and regularized fine-tuning method (GRFT) that updates the rows or columns of the weight matrix.<n> GRFT achieves state-of-the-art performance, surpassing existing methods such as GPS, Adapter Tuning, and LoRA.
  arXiv  Detail & Related papers  (2025-06-14T14:41:03Z)
• 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)
• Scaling Exponents Across Parameterizations and Optimizers [94.54718325264218]
  We propose a new perspective on parameterization by investigating a key assumption in prior work. Our empirical investigation includes tens of thousands of models trained with all combinations of threes. We find that the best learning rate scaling prescription would often have been excluded by the assumptions in prior work.
  arXiv  Detail & Related papers  (2024-07-08T12:32:51Z)
• Parameter-Efficient Fine-Tuning without Introducing New Latency [7.631596468553607]
  We introduce a novel adapter technique that directly applies the adapter to pre-trained parameters instead of the hidden representation. Our proposed method attains a new state-of-the-art outcome in terms of both performance and storage efficiency, storing only 0.03% parameters of full fine-tuning.
  arXiv  Detail & Related papers  (2023-05-26T08:44:42Z)
• AdaLoRA: Adaptive Budget Allocation for Parameter-Efficient Fine-Tuning [143.23123791557245]
  Fine-tuning large pre-trained language models on downstream tasks has become an important paradigm in NLP. We propose AdaLoRA, which adaptively allocates the parameter budget among weight matrices according to their importance score. We conduct extensive experiments with several pre-trained models on natural language processing, question answering, and natural language generation to validate the effectiveness of AdaLoRA.
  arXiv  Detail & Related papers  (2023-03-18T22:36:25Z)
• Sensitivity-Aware Visual Parameter-Efficient Fine-Tuning [91.5113227694443]
  We propose a novel visual. sensuous-aware fine-Tuning (SPT) scheme. SPT allocates trainable parameters to task-specific important positions. Experiments on a wide range of downstream recognition tasks show that our SPT is complementary to the existing PEFT methods.
  arXiv  Detail & Related papers  (2023-03-15T12:34:24Z)
• Scaling & Shifting Your Features: A New Baseline for Efficient Model Tuning [126.84770886628833]
  Existing finetuning methods either tune all parameters of the pretrained model (full finetuning) or only tune the last linear layer (linear probing) We propose a new parameter-efficient finetuning method termed as SSF, representing that researchers only need to Scale and Shift the deep Features extracted by a pre-trained model to catch up with the performance full finetuning.
  arXiv  Detail & Related papers  (2022-10-17T08:14:49Z)
• 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)

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.