# Revisiting Visual Representation Enhancement of VLMs via Kernel Canonical Correlation Analysis

arXiv: https://arxiv.org/abs/2610.02718 · HTML: https://mlsanity.com/paper/2610.02718/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Prompt learning and adaptation of CLIP, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 18 of its 25 references that are in the map.
- First posted on arXiv 2 Oct 2026.

## Placed in

- Multimodal LLMs / vision-language models / [Prompt learning and adaptation of CLIP](https://mlsanity.com/mllm/prompt-learning-and-adaptation-of-clip/)

## Builds on (most cited first, 18 of 18)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Barlow Twins: Self-Supervised Learning via Redundancy Reduction](https://mlsanity.com/paper/2103.03230/) (2021)
- [Microsoft COCO Captions: Data Collection and Evaluation Server](https://mlsanity.com/paper/1504.00325/) (2015)
- [VICReg: Variance-Invariance-Covariance Regularization for Self-Supervised Learning](https://mlsanity.com/paper/2105.04906/) (2021)
- [Eyes Wide Shut? Exploring the Visual Shortcomings of Multimodal LLMs](https://mlsanity.com/paper/2401.06209/) (2024)
- [Language-driven Semantic Segmentation](https://mlsanity.com/paper/2201.03546/) (2022)
- [RegionCLIP: Region-based Language-Image Pretraining](https://mlsanity.com/paper/2112.09106/) (2021)
- [Simple Open-Vocabulary Object Detection with Vision Transformers](https://mlsanity.com/paper/2205.06230/) (2022)
- [CLIP-DINOiser: Teaching CLIP a few DINO tricks for open-vocabulary semantic segmentation](https://mlsanity.com/paper/2312.12359/) (2023)
- [FG-CLIP: Fine-Grained Visual and Textual Alignment](https://mlsanity.com/paper/2505.05071/) (2025)
- [Improved Zero-Shot Classification by Adapting VLMs with Text Descriptions](https://mlsanity.com/paper/2401.02460/) (2024)
- [Talking to DINO: Bridging Self-Supervised Vision Backbones with Language for Open-Vocabulary Segmentation](https://mlsanity.com/paper/2411.19331/) (2024)
- [Diffusion Feedback Helps CLIP See Better](https://mlsanity.com/paper/2407.20171/) (2024)
- [Deep Generalized Canonical Correlation Analysis](https://mlsanity.com/paper/1702.02519/) (2017)
- [Kernel-based Unsupervised Embedding Alignment for Enhanced Visual Representation in Vision-language Models](https://mlsanity.com/paper/2506.02557/) (2025)
- [Unconstrained Stochastic CCA: Unifying Multiview and Self-Supervised Learning](https://mlsanity.com/paper/2310.01012/) (2023)
- [KALE: Kernel Alignment with Loss Equilibration for Stable CLIP-DINOv2 Alignment at Web Scale](https://mlsanity.com/paper/2607.18885/) (2026)

## Abstract (arXiv)

Vision-language models such as CLIP exhibit strong semantic generalization, but remain limited in fine-grained visual perception. A recent work named KUEA presents a natural remedy by finetuning the image encoder under the supervision of the vision-centric DINOv2 to align their kernel matrices element-wisely, while regularizing the embeddings to remain close to the pretrained visual encoder for preserving image-text semantics in CLIP. However, we show that diminishing the role of the alignment loss to DINOv2 does not necessarily degrade its fine-grained visual performance, suggesting that the kernel-matrix discrepancy may be insufficient for further visual representation enhancement, motivating us to revisit the alignment formulation. In this work, we present a novel perspective to characterize representation alignment on feature subspaces through Kernel Canonical Correlation Analysis (KCCA), which maximizes the projection correlations. In optimization, we derive an efficient end-to-end training scheme upon KKT conditions, avoiding the eigenvalue problem in KCCA. Further, we extend our method into a 3-view formulation, i.e., 3vKCCA, in which the projections from the pretrained text encoder are also incorporated under a unified optimization framework for joint alignment. With CLIP ViT-L/14 on ImageNet-1K, our 3vKCCA improves the MMVP-VLM accuracy from 17.8 to 25.9, substantially outperforming the existing methods, and meanwhile maintains zero-shot image--text retrieval performance.