# Exploring Negatives in Contrastive Learning for Unpaired Image-to-Image Translation arXiv: https://arxiv.org/abs/2204.11018 ยท HTML: https://mlsanity.com/paper/2204.11018/ ## Key facts - Cited by 6 papers in the map (data through 14 Oct 2026). - Builds on 38 of its 47 references that are in the map. - First posted on arXiv 23 Apr 2022. ## Builds on (most cited first, 30 of 38) - [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020) - [Representation Learning with Contrastive Predictive Coding](https://mlsanity.com/paper/1807.03748/) (2018) - [Momentum Contrast for Unsupervised Visual Representation Learning](https://mlsanity.com/paper/1911.05722/) (2019) - [Image-to-Image Translation with Conditional Adversarial Networks](https://mlsanity.com/paper/1611.07004/) (2016) - [The Cityscapes Dataset for Semantic Urban Scene Understanding](https://mlsanity.com/paper/1604.01685/) (2016) - [Analyzing and Improving the Image Quality of StyleGAN](https://mlsanity.com/paper/1912.04958/) (2019) - [Understanding Contrastive Representation Learning through Alignment and Uniformity on the Hypersphere](https://mlsanity.com/paper/2005.10242/) (2020) - [Semantic Image Synthesis with Spatially-Adaptive Normalization](https://mlsanity.com/paper/1903.07291/) (2019) - [Learning deep representations by mutual information estimation and maximization](https://mlsanity.com/paper/1808.06670/) (2018) - [High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs](https://mlsanity.com/paper/1711.11585/) (2017) - [Multimodal Unsupervised Image-to-Image Translation](https://mlsanity.com/paper/1804.04732/) (2018) - [Encoding in Style: a StyleGAN Encoder for Image-to-Image Translation](https://mlsanity.com/paper/2008.00951/) (2020) - [Which Training Methods for GANs do actually Converge?](https://mlsanity.com/paper/1801.04406/) (2018) - [Data-Efficient Image Recognition with Contrastive Predictive Coding](https://mlsanity.com/paper/1905.09272/) (2019) - [Contrastive Learning for Unpaired Image-to-Image Translation](https://mlsanity.com/paper/2007.15651/) (2020) - [Unsupervised Image-to-Image Translation Networks](https://mlsanity.com/paper/1703.00848/) (2017) - [On Variational Bounds of Mutual Information](https://mlsanity.com/paper/1905.06922/) (2019) - [DualGAN: Unsupervised Dual Learning for Image-to-Image Translation](https://mlsanity.com/paper/1704.02510/) (2017) - [Dilated Residual Networks](https://mlsanity.com/paper/1705.09914/) (2017) - [U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation](https://mlsanity.com/paper/1907.10830/) (2019) - [RankSRGAN: Generative Adversarial Networks with Ranker for Image Super-Resolution](https://mlsanity.com/paper/1908.06382/) (2019) - [Image-to-Image Translation: Methods and Applications](https://mlsanity.com/paper/2101.08629/) (2021) - [Geometry-Consistent Generative Adversarial Networks for One-Sided Unsupervised Domain Mapping](https://mlsanity.com/paper/1809.05852/) (2018) - [One-Sided Unsupervised Domain Mapping](https://mlsanity.com/paper/1706.00826/) (2017) - [The Spatially-Correlative Loss for Various Image Translation Tasks](https://mlsanity.com/paper/2104.00854/) (2021) - [Dual Contrastive Learning for Unsupervised Image-to-Image Translation](https://mlsanity.com/paper/2104.07689/) (2021) - [Image-to-image Translation via Hierarchical Style Disentanglement](https://mlsanity.com/paper/2103.01456/) (2021) - [Disentangled Cycle Consistency for Highly-realistic Virtual Try-On](https://mlsanity.com/paper/2103.09479/) (2021) - [Rethinking the Truly Unsupervised Image-to-Image Translation](https://mlsanity.com/paper/2006.06500/) (2020) - [Frequency Domain Image Translation: More Photo-realistic, Better Identity-preserving](https://mlsanity.com/paper/2011.13611/) (2020) ## Built on by (most cited first, 6 of 6) - [Criteria Comparative Learning for Real-scene Image Super-Resolution](https://mlsanity.com/paper/2207.12767/) (2022) - [ReGeneration Learning of Diffusion Models with Rich Prompts for Zero-Shot Image Translation](https://mlsanity.com/paper/2305.04651/) (2023) - [Multi-cropping Contrastive Learning and Domain Consistency for Unsupervised Image-to-Image Translation](https://mlsanity.com/paper/2304.12235/) (2023) - [Diff-Oracle: Deciphering Oracle Bone Scripts with Controllable Diffusion Model](https://mlsanity.com/paper/2312.13631/) (2023) - [Unveiling the Underwater World: CLIP Perception Model-Guided Underwater Image Enhancement](https://mlsanity.com/paper/2507.06234/) (2025) - [An Analysis for Image-to-Image Translation and Style Transfer](https://mlsanity.com/paper/2408.06000/) (2024) ## Abstract (arXiv) Unpaired image-to-image translation aims to find a mapping between the source domain and the target domain. To alleviate the problem of the lack of supervised labels for the source images, cycle-consistency based methods have been proposed for image structure preservation by assuming a reversible relationship between unpaired images. However, this assumption only uses limited correspondence between image pairs. Recently, contrastive learning (CL) has been used to further investigate the image correspondence in unpaired image translation by using patch-based positive/negative learning. Patch-based contrastive routines obtain the positives by self-similarity computation and recognize the rest patches as negatives. This flexible learning paradigm obtains auxiliary contextualized information at a low cost. As the negatives own an impressive sample number, with curiosity, we make an investigation based on a question: are all negatives necessary for feature contrastive learning? Unlike previous CL approaches that use negatives as much as possible, in this paper, we study the negatives from an information-theoretic perspective and introduce a new negative Pruning technology for Unpaired image-to-image Translation (PUT) by sparsifying and ranking the patches. The proposed algorithm is efficient, flexible and enables the model to learn essential information between corresponding patches stably. By putting quality over quantity, only a few negative patches are required to achieve better results. Lastly, we validate the superiority, stability, and versatility of our model through comparative experiments.