Preserve and Sculpt: Manifold-Aligned Fine-tuning of Vision-Language Models for Few-Shot Learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Dexia, Zhu, Qianjie, Li, Weibing, Yu, Yue, Zhang, Tong, Wang, Ruixuan
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913996259459072
author Chen, Dexia
Zhu, Qianjie
Li, Weibing
Yu, Yue
Zhang, Tong
Wang, Ruixuan
author_facet Chen, Dexia
Zhu, Qianjie
Li, Weibing
Yu, Yue
Zhang, Tong
Wang, Ruixuan
contents Pretrained vision-language models (VLMs), such as CLIP, have shown remarkable potential in few-shot image classification and led to numerous effective transfer learning strategies. These methods leverage the pretrained knowledge of VLMs to enable effective domain adaptation while mitigating overfitting through parameter-efficient tuning or instance-based consistency constraints. However, such regularizations often neglect the geometric structure of data distribution, which may lead to distortion of the overall semantic representation. To overcome this limitation, we propose a novel fine-tuning method, Manifold-Preserving and Sculpting Tuning (MPS-Tuning). Regarding the data distribution in feature space as a semantic manifold, MPS-Tuning explicitly constrains the intrinsic geometry of this manifold while further sculpting it to enhance class separability. Specifically, MPS-Tuning preserves both macroscopic and microscopic topological structures of the original manifold by aligning Gram matrices of features before and after fine-tuning. Theoretically, this constraint is shown to approximate an upper bound of the Gromov-Wasserstein distance. Furthermore, features from the image and text modalities are paired, and pairwise similarities are optimized to enhance the manifold's class discriminability. Extensive experiments demonstrate that MPS-Tuning significantly improves model performance while effectively preserving the structure of the semantic manifold. The code will be released.
format Preprint
id arxiv_https___arxiv_org_abs_2508_12877
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Preserve and Sculpt: Manifold-Aligned Fine-tuning of Vision-Language Models for Few-Shot Learning
Chen, Dexia
Zhu, Qianjie
Li, Weibing
Yu, Yue
Zhang, Tong
Wang, Ruixuan
Computer Vision and Pattern Recognition
Pretrained vision-language models (VLMs), such as CLIP, have shown remarkable potential in few-shot image classification and led to numerous effective transfer learning strategies. These methods leverage the pretrained knowledge of VLMs to enable effective domain adaptation while mitigating overfitting through parameter-efficient tuning or instance-based consistency constraints. However, such regularizations often neglect the geometric structure of data distribution, which may lead to distortion of the overall semantic representation. To overcome this limitation, we propose a novel fine-tuning method, Manifold-Preserving and Sculpting Tuning (MPS-Tuning). Regarding the data distribution in feature space as a semantic manifold, MPS-Tuning explicitly constrains the intrinsic geometry of this manifold while further sculpting it to enhance class separability. Specifically, MPS-Tuning preserves both macroscopic and microscopic topological structures of the original manifold by aligning Gram matrices of features before and after fine-tuning. Theoretically, this constraint is shown to approximate an upper bound of the Gromov-Wasserstein distance. Furthermore, features from the image and text modalities are paired, and pairwise similarities are optimized to enhance the manifold's class discriminability. Extensive experiments demonstrate that MPS-Tuning significantly improves model performance while effectively preserving the structure of the semantic manifold. The code will be released.
title Preserve and Sculpt: Manifold-Aligned Fine-tuning of Vision-Language Models for Few-Shot Learning
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2508.12877