Mind the Discriminability Trap in Source-Free Cross-domain Few-shot Learning

• URL: http://arxiv.org/abs/2603.13341v1
• Date: Sat, 07 Mar 2026 03:59:23 GMT
• Title: Mind the Discriminability Trap in Source-Free Cross-domain Few-shot Learning
• Abstract summary: Source-Free Cross-Domain Few-Shot Learning focuses on fine-tuning with limited training data from target domains.<n>We find that improving visual discriminability actually suppresses VLMs' performance.<n>We propose an approach to address this problem: first, we perturb the visual learning to guide the model to focus on the cross-modal alignment.
• Score: 30.80780619903459
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Source-Free Cross-Domain Few-Shot Learning (SF-CDFSL) focuses on fine-tuning with limited training data from target domains (e.g., medical or satellite images), where Vision-Language Models (VLMs) such as CLIP and SigLIP have shown promising results. Current works in traditional visual models suggest that improving visual discriminability enhances performance. However, in VLM-based SF-CDFSL tasks, we find that \textbf{strengthening visual-modal discriminability actually suppresses VLMs' performance}. In this paper, we aim to delve into this phenomenon for an interpretation and a solution. By both theoretical and experimental proofs, our study reveals that fine-tuning with the typical cross-entropy loss ($\mathcal{L}_{\mathrm{vlm}}$) inherently includes a visual learning part and a cross-modal learning part, where the cross-modal part is crucial for rectifying the heavily disrupted modality misalignment in SF-CDFSL. However, we find that the visual learning essentially acts as a shortcut that encourages the model to reduce $\mathcal{L}_{\mathrm{vlm}}$ without considering the cross-modal part, therefore hindering the cross-modal alignment and harming the performance. Based on this interpretation, we further propose an approach to address this problem: first, we perturb the visual learning to guide the model to focus on the cross-modal alignment. Then, we use the visual-text semantic relationships to gradually align the visual and textual modalities during the fine-tuning. Extensive experiments on various settings, backbones (CLIP, SigLip, PE-Core), and tasks (4 CDFSL datasets and 11 FSL datasets) show that we consistently set new state-of-the-art results. Code is available at https://github.com/zhenyuZ-HUST/CVPR26-Mind-the-Discriminability-Trap.

Related papers

• Reviving In-domain Fine-tuning Methods for Source-Free Cross-domain Few-shot Learning [19.4465469479799]
  Cross-Domain Few-Shot Learning aims to adapt large-scale pretrained models to specialized target domains with limited samples.<n>We find adapter-based methods consistently outperform prompt-based ones, contrary to in-domain scenarios.<n>We propose Semantic Probe, a plug-and-play attention rectification framework for both adapter- and prompt-based methods.
  arXiv  Detail & Related papers  (2026-05-12T07:20:23Z)
• Interpretable Cross-Domain Few-Shot Learning with Rectified Target-Domain Local Alignment [19.113214017897118]
  Cross-Domain Few-Shot Learning adapts models trained with large-scale general data (source domain) to downstream target domains with only scarce training data.<n>We find that CLIP models can hardly focus on fine-grained visual cues for interpretable recognition.<n>To address this problem, due to the lack of supervision in aligning local visual features and text semantics, we turn to self-supervision information.
  arXiv  Detail & Related papers  (2026-03-18T12:20:21Z)
• Causal Disentanglement and Cross-Modal Alignment for Enhanced Few-Shot Learning [11.752632557524969]
  Causal CLIP Adapter (CCA) is a novel framework that explicitly disentangles visual features extracted from CLIP.<n>Our method consistently outperforms state-of-the-art approaches in terms of few-shot performance and robustness to distributional shifts.
  arXiv  Detail & Related papers  (2025-08-05T05:30:42Z)
• Interpretable Zero-Shot Learning with Locally-Aligned Vision-Language Model [56.573203512455706]
  Large-scale vision-language models (VLMs) have achieved remarkable success in zero-shot learning (ZSL) by leveraging large-scale visual-text pair datasets.<n>One approach to address this issue is to develop interpretable models by integrating language.<n>We propose LaZSL, a locally-aligned vision-language model for interpretable ZSL.
  arXiv  Detail & Related papers  (2025-06-30T13:14:46Z)
• Mitigating Hallucination for Large Vision Language Model by Inter-Modality Correlation Calibration Decoding [66.06337890279839]
  Large vision-language models (LVLMs) have shown remarkable capabilities in visual-language understanding for downstream multi-modal tasks.<n>LVLMs still suffer from generating hallucinations in complex generation tasks, leading to inconsistencies between visual inputs and generated content.<n>We propose an Inter-Modality Correlation Decoding (IMCCD) method to mitigate hallucinations in LVLMs in a training-free manner.
  arXiv  Detail & Related papers  (2025-01-03T17:56:28Z)
• COSMOS: Cross-Modality Self-Distillation for Vision Language Pre-training [49.2684130383925]
  We propose COSMOS: CrOSs-MOdality Self-distillation for vision-language pre-training.<n>It integrates a novel text-cropping strategy and cross-attention module into a self-supervised learning framework.<n>It consistently outperforms previous strong baselines on various zero-shot downstream tasks.
  arXiv  Detail & Related papers  (2024-12-02T18:56:06Z)
• X-VILA: Cross-Modality Alignment for Large Language Model [91.96081978952283]
  X-VILA is an omni-modality model designed to extend the capabilities of large language models (LLMs) by incorporating image, video, and audio modalities. We propose a visual alignment mechanism with a visual embedding highway module to address the problem of visual information loss. X-VILA exhibits proficiency in any-to-any modality conversation, surpassing previous approaches by large margins.
  arXiv  Detail & Related papers  (2024-05-29T17:59:58Z)
• Continual Vision-Language Representation Learning with Off-Diagonal Information [112.39419069447902]
  Multi-modal contrastive learning frameworks like CLIP typically require a large amount of image-text samples for training. This paper discusses the feasibility of continual CLIP training using streaming data.
  arXiv  Detail & Related papers  (2023-05-11T08:04:46Z)
• Black Box Few-Shot Adaptation for Vision-Language models [41.49584259596654]
  Vision-Language (V-L) models trained with contrastive learning to align the visual and language modalities have been shown to be strong few-shot learners. We describe a black-box method for V-L few-shot adaptation that operates on pre-computed image and text features. We propose Linear Feature Alignment (LFA), a simple linear approach for V-L re-alignment in the target domain.
  arXiv  Detail & Related papers  (2023-04-04T12:42:29Z)
• FREE: Feature Refinement for Generalized Zero-Shot Learning [86.41074134041394]
  Generalized zero-shot learning (GZSL) has achieved significant progress, with many efforts dedicated to overcoming the problems of visual-semantic domain gap and seen-unseen bias. Most existing methods directly use feature extraction models trained on ImageNet alone, ignoring the cross-dataset bias between ImageNet and GZSL benchmarks. We propose a simple yet effective GZSL method, termed feature refinement for generalized zero-shot learning (FREE) to tackle the above problem.
  arXiv  Detail & Related papers  (2021-07-29T08:11:01Z)

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.