Linearization Explains Fine-Tuning in Large Language Models

• URL: http://arxiv.org/abs/2602.08239v1
• Date: Mon, 09 Feb 2026 03:27:58 GMT
• Title: Linearization Explains Fine-Tuning in Large Language Models
• Authors: Zahra Rahimi Afzal, Tara Esmaeilbeig, Mojtaba Soltanalian, Mesrob I. Ohannessian,
• Abstract summary: We show that fine-tuning dynamics become equivalent to learning with the positive-definite neural tangent kernel (NTK)<n>When linearization is a good model, our findings reveal a strong correlation between the eigenvalue spectrum of the NTK and the performance of model adaptation.
• Score: 13.157568306905885
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Parameter-Efficient Fine-Tuning (PEFT) is a popular class of techniques that strive to adapt large models in a scalable and resource-efficient manner. Yet, the mechanisms underlying their training performance and generalization remain underexplored. In this paper, we provide several insights into such fine-tuning through the lens of linearization. Fine-tuned models are often implicitly encouraged to remain close to the pretrained model. By making this explicit, using an Euclidean distance inductive bias in parameter space, we show that fine-tuning dynamics become equivalent to learning with the positive-definite neural tangent kernel (NTK). We specifically analyze how close the fully linear and the linearized fine-tuning optimizations are, based on the strength of the regularization. This allows us to be pragmatic about how good a model linearization is when fine-tuning large language models (LLMs). When linearization is a good model, our findings reveal a strong correlation between the eigenvalue spectrum of the NTK and the performance of model adaptation. Motivated by this, we give spectral perturbation bounds on the NTK induced by the choice of layers selected for fine-tuning. We empirically validate our theory on Low Rank Adaptation (LoRA) on LLMs. These insights not only characterize fine-tuning but also have the potential to enhance PEFT techniques, paving the way to better informed and more nimble adaptation in LLMs.

Related papers

• ODELoRA: Training Low-Rank Adaptation by Solving Ordinary Differential Equations [54.886931928255564]
  Low-rank adaptation (LoRA) has emerged as a widely adopted parameter-efficient fine-tuning method in deep transfer learning.<n>We propose a novel continuous-time optimization dynamic for LoRA factor matrices in the form of an ordinary differential equation (ODE)<n>We show that ODELoRA achieves stable feature learning, a property that is crucial for training deep neural networks at different scales of problem dimensionality.
  arXiv  Detail & Related papers  (2026-02-07T10:19:36Z)
• $α$-LoRA: Effective Fine-Tuning via Base Model Rescaling [41.58663029548425]
  We introduce a new class of re parameterization methods for transfer learning, designed to enhance the ability generalization of fine-tuned models.<n>We establish the effectiveness of our approach in a high-dimensional binary classification setting using tools from Random Matrix Theory, and further validate our theoretical findings through more realistic experiments.
  arXiv  Detail & Related papers  (2025-10-24T11:19:33Z)
• Implicit Reward as the Bridge: A Unified View of SFT and DPO Connections [65.36449542323277]
  We present a unified theoretical framework bridgingSupervised Fine-Tuning (SFT) and preference learning in Large Language Model (LLM) post-training.<n>We propose a simple yet effective learning rate reduction approach that yields significant performance improvements.
  arXiv  Detail & Related papers  (2025-06-15T05:42:29Z)
• NIMO: a Nonlinear Interpretable MOdel [5.128077543874915]
  NIMO is a framework that combines inherent interpretability with the expressive power of neural networks.<n>We show that our model can provide faithful and intelligible feature effects while maintaining good predictive performance.
  arXiv  Detail & Related papers  (2025-06-05T14:02:55Z)
• Weight Spectra Induced Efficient Model Adaptation [54.8615621415845]
  Fine-tuning large-scale foundation models incurs prohibitive computational costs.<n>We show that fine-tuning predominantly amplifies the top singular values while leaving the remainder largely intact.<n>We propose a novel method that leverages learnable rescaling of top singular directions.
  arXiv  Detail & Related papers  (2025-05-29T05:03:29Z)
• SMILE: Zero-Shot Sparse Mixture of Low-Rank Experts Construction From Pre-Trained Foundation Models [85.67096251281191]
  We present an innovative approach to model fusion called zero-shot Sparse MIxture of Low-rank Experts (SMILE) construction. SMILE allows for the upscaling of source models into an MoE model without extra data or further training. We conduct extensive experiments across diverse scenarios, such as image classification and text generation tasks, using full fine-tuning and LoRA fine-tuning.
  arXiv  Detail & Related papers  (2024-08-19T17:32:15Z)
• Spectrum-Aware Parameter Efficient Fine-Tuning for Diffusion Models [73.88009808326387]
  We propose a novel spectrum-aware adaptation framework for generative models. Our method adjusts both singular values and their basis vectors of pretrained weights. We introduce Spectral Ortho Decomposition Adaptation (SODA), which balances computational efficiency and representation capacity.
  arXiv  Detail & Related papers  (2024-05-31T17:43:35Z)
• Understanding Linear Probing then Fine-tuning Language Models from NTK Perspective [32.01426831450348]
  The two-stage fine-tuning (FT) method, linear probing (LP) then fine-tuning (LP-FT), outperforms linear probing and FT alone. We analyze the training dynamics of LP-FT for classification tasks on the basis of the neural tangent kernel (NTK) theory. Our study demonstrates the effectiveness of LP-FT for fine-tuning language models.
  arXiv  Detail & Related papers  (2024-05-27T01:31:40Z)
• Edge-Efficient Deep Learning Models for Automatic Modulation Classification: A Performance Analysis [0.7428236410246183]
  We investigate optimized convolutional neural networks (CNNs) developed for automatic modulation classification (AMC) of wireless signals. We propose optimized models with the combinations of these techniques to fuse the complementary optimization benefits. The experimental results show that the proposed individual and combined optimization techniques are highly effective for developing models with significantly less complexity.
  arXiv  Detail & Related papers  (2024-04-11T06:08:23Z)
• On the Emergence of Cross-Task Linearity in the Pretraining-Finetuning Paradigm [47.55215041326702]
  We discover an intriguing linear phenomenon in models that are from a common pretrained checkpoint and finetuned on different tasks, termed as Cross-Task Linearity (CTL) We show that if we linearly interpolate the weights of two finetuned models, the features in the weight-interpolated model are often approximately equal to the linearities of features in two finetuned models at each layer. We conjecture that in the pretraining-finetuning paradigm, neural networks approximately function as linear maps, mapping from the parameter space to the feature space.
  arXiv  Detail & Related papers  (2024-02-06T03:28:36Z)
• LQF: Linear Quadratic Fine-Tuning [114.3840147070712]
  We present the first method for linearizing a pre-trained model that achieves comparable performance to non-linear fine-tuning. LQF consists of simple modifications to the architecture, loss function and optimization typically used for classification.
  arXiv  Detail & Related papers  (2020-12-21T06:40:20Z)

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.