# LaCViT: A Label-aware Contrastive Fine-tuning Framework for Vision Transformers

arXiv: https://arxiv.org/abs/2303.18013 · HTML: https://mlsanity.com/paper/2303.18013/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 32 references that are in the map.
- First posted on arXiv 31 Mar 2023.

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

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [Training data-efficient image transformers & distillation through attention](https://mlsanity.com/paper/2012.12877/) (2020)
- [Supervised Contrastive Learning](https://mlsanity.com/paper/2004.11362/) (2020)
- [SimMIM: A Simple Framework for Masked Image Modeling](https://mlsanity.com/paper/2111.09886/) (2021)
- [Learning Imbalanced Datasets with Label-Distribution-Aware Margin Loss](https://mlsanity.com/paper/1906.07413/) (2019)
- [Image Transformer](https://mlsanity.com/paper/1802.05751/) (2018)
- [data2vec: A General Framework for Self-supervised Learning in Speech, Vision and Language](https://mlsanity.com/paper/2202.03555/) (2022)
- [LeViT: a Vision Transformer in ConvNet's Clothing for Faster Inference](https://mlsanity.com/paper/2104.01136/) (2021)
- [Large-Margin Softmax Loss for Convolutional Neural Networks](https://mlsanity.com/paper/1612.02295/) (2016)
- [ViTAE: Vision Transformer Advanced by Exploring Intrinsic Inductive Bias](https://mlsanity.com/paper/2106.03348/) (2021)
- [Scaling Autoregressive Video Models](https://mlsanity.com/paper/1906.02634/) (2019)
- [ConvNets vs. Transformers: Whose Visual Representations are More Transferable?](https://mlsanity.com/paper/2108.05305/) (2021)
- [Robust Data2vec: Noise-robust Speech Representation Learning for ASR by Combining Regression and Improved Contrastive Learning](https://mlsanity.com/paper/2210.15324/) (2022)
- [Cross-Entropy Loss and Low-Rank Features Have Responsibility for Adversarial Examples](https://mlsanity.com/paper/1901.08360/) (2019)
- [ESCL: Equivariant Self-Contrastive Learning for Sentence Representations](https://mlsanity.com/paper/2303.05143/) (2023)

## Built on by (most cited first, 5 of 5)

- [RoboLLM: Robotic Vision Tasks Grounded on Multimodal Large Language Models](https://mlsanity.com/paper/2310.10221/) (2023)
- [CFIR: Fast and Effective Long-Text To Image Retrieval for Large Corpora](https://mlsanity.com/paper/2402.15276/) (2024)
- [CLCE: An Approach to Refining Cross-Entropy and Contrastive Learning for Optimized Learning Fusion](https://mlsanity.com/paper/2402.14551/) (2024)
- [Understanding and Mitigating Human-Labelling Errors in Supervised Contrastive Learning](https://mlsanity.com/paper/2403.06289/) (2024)
- [GCAM: Gaussian and causal-attention model of food fine-grained recognition](https://mlsanity.com/paper/2403.12109/) (2024)

## Abstract (arXiv)

Vision Transformers (ViTs) have emerged as popular models in computer vision, demonstrating state-of-the-art performance across various tasks. This success typically follows a two-stage strategy involving pre-training on large-scale datasets using self-supervised signals, such as masked random patches, followed by fine-tuning on task-specific labeled datasets with cross-entropy loss. However, this reliance on cross-entropy loss has been identified as a limiting factor in ViTs, affecting their generalization and transferability to downstream tasks. Addressing this critical challenge, we introduce a novel Label-aware Contrastive Training framework, LaCViT, which significantly enhances the quality of embeddings in ViTs. LaCViT not only addresses the limitations of cross-entropy loss but also facilitates more effective transfer learning across diverse image classification tasks. Our comprehensive experiments on eight standard image classification datasets reveal that LaCViT statistically significantly enhances the performance of three evaluated ViTs by up-to 10.78% under Top-1 Accuracy.