Research map / Vision-language and multimodal / Multimodal LLMs
CLIP-Adapter: Better Vision-Language Models with Feature Adapters
Peng Gao, Shijie Geng, Renrui Zhang, Teli Ma, Rongyao Fang, Yongfeng Zhang, Hongsheng Li, Yu Qiao
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add residual feature adapter on frozen CLIP image features | Zero-shot CLIP RN50 · ImageNet (16-shot) | accuracy: 55.41% -> 61.33% | Improved |
| replace CoOp prompt tuning with CLIP-Adapter | CoOp RN50 · ImageNet (16-shot) | accuracy: 60.46% -> 61.33% | Improved |
| use CLIP-Adapter instead of CoOp for training time | CoOp RN50 · ImageNet (16-shot) | train time: 14h 40min -> 50min | Improved |
| add residual connection (alpha 0.6 instead of 1.0) (ablation) | CLIP-Adapter · DTD (16-shot) | accuracy: 63.79 -> 66.06 | Improved |
| add residual connection (alpha 0.2 instead of 1.0) (ablation) | CLIP-Adapter · ImageNet (16-shot) | accuracy: 59.05 -> 61.33 | Improved |
| insert adapter only at end of encoder instead of all layers (ablation) | CLIP-Adapter ViT-B/16 · ImageNet (16-shot) | accuracy: 67.67 -> 70.88 | Improved |
| insert adapter at layer 10 instead of layer 12 (ablation) | CLIP-Adapter ViT-B/16 · ImageNet (16-shot) | accuracy: 70.88 -> 71.85 | Mixed |
| set bottleneck dimension to D/4 instead of D (ablation) | CLIP-Adapter · ImageNet (16-shot) | accuracy: 59.78 -> 61.33 | Improved |
| shrink bottleneck dimension from D/4 to D/32 (ablation) | CLIP-Adapter · ImageNet (16-shot) | accuracy: 61.33 -> 59.45 | Hurt |
| also finetune CLIP visual encoder (ablation) | CLIP-Adapter · ImageNet (16-shot) | accuracy: 61.33% -> 60.07% | Hurt |
| also finetune both CLIP encoders (ablation) | CLIP-Adapter · ImageNet (16-shot) | accuracy: 61.33% -> 52.78% | Hurt |
| use ensemble of 7 hard prompts (ablation) | CLIP-Adapter · ImageNet (16-shot) | accuracy: 61.33 -> 61.68 | Improved |
Later papers that compared with this method
8 later accepted papers report CLIP-Adapter as a baseline in their results tables; 8 of them report a better number on most of the results they share with it.
- Distributional Vision-Language Alignment by Cauchy-Schwarz Divergence (ICLR 2026 · better on 1 of 1 shared result)
- Exploring Cross-Modal Flows for Few-Shot Learning (ICLR 2026 · better on 1 of 1 shared result)
- Reproducible Vision-Language Models Meet Concepts Out of Pre-Training (CVPR 2025 · better on 4 of 4 shared results)
- Enhance Vision-Language Alignment with Noise (AAAI 2025 · better on 1 of 1 shared result)
- Skip Tuning: Pre-trained Vision-Language Models are Effective and Efficient Adapters Themselves (CVPR 2025 · better on 1 of 1 shared result)
- Robust Calibration of Large Vision-Language Adapters (ECCV 2024 · better on 3 of 3 shared results)
- Bridging the Pathology Domain Gap: Efficiently Adapting CLIP for Pathology Image Analysis with Limited Labeled Data (ECCV 2024 · better on 1 of 1 shared result)
- Regularized Mask Tuning: Uncovering Hidden Knowledge in Pre-Trained Vision-Language Models (ICCV 2023 · better on 1 of 1 shared result)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Post-pre-training for Modality Alignment in Vision-Language Foundation Models (CVPR 2025 · similarity 0.91)
- A Hard-to-Beat Baseline for Training-free CLIP-based Adaptation (ICLR 2024 · similarity 0.91)
- Learning to Prompt with Text Only Supervision for Vision-Language Models (AAAI 2025 · similarity 0.90)
- IPO: Interpretable Prompt Optimization for Vision-Language Models (NeurIPS 2024 · similarity 0.90)
- Adaptive Parameter Selection for Tuning Vision-Language Models (CVPR 2025 · similarity 0.90)
- ProKeR: A Kernel Perspective on Few-Shot Adaptation of Large Vision-Language Models (CVPR 2025 · similarity 0.90)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
