# Learning Attention as Disentangler for Compositional Zero-shot Learning

arXiv: https://arxiv.org/abs/2303.15111 · HTML: https://mlsanity.com/paper/2303.15111/

## Key facts

- Cited by 29 papers in the map (data through 14 Oct 2026).
- Builds on 28 of its 52 references that are in the map.
- First posted on arXiv 27 Mar 2023.

## Builds on (most cited first, 28 of 28)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Emerging Properties in Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.14294/) (2021)
- [Generative Adversarial Networks](https://mlsanity.com/paper/2203.00667/) (2022)
- [Distributed Representations of Words and Phrases and their Compositionality](https://mlsanity.com/paper/1310.4546/) (2013)
- [Non-local Neural Networks](https://mlsanity.com/paper/1711.07971/) (2017)
- [Do Vision Transformers See Like Convolutional Neural Networks?](https://mlsanity.com/paper/2108.08810/) (2021)
- [Feature Generating Networks for Zero-Shot Learning](https://mlsanity.com/paper/1712.00981/) (2017)
- [Zero-Shot Learning -- The Good, the Bad and the Ugly](https://mlsanity.com/paper/1703.04394/) (2017)
- [Zero-Shot Learning Through Cross-Modal Transfer](https://mlsanity.com/paper/1301.3666/) (2013)
- [An Empirical Study and Analysis of Generalized Zero-Shot Learning for Object Recognition in the Wild](https://mlsanity.com/paper/1605.04253/) (2016)
- [Task-Driven Modular Networks for Zero-Shot Compositional Learning](https://mlsanity.com/paper/1905.05908/) (2019)
- [Learning Graph Embeddings for Compositional Zero-shot Learning](https://mlsanity.com/paper/2102.01987/) (2021)
- [Open World Compositional Zero-Shot Learning](https://mlsanity.com/paper/2101.12609/) (2021)
- [Learning to Predict Visual Attributes in the Wild](https://mlsanity.com/paper/2106.09707/) (2021)
- [Symmetry and Group in Attribute-Object Compositions](https://mlsanity.com/paper/2004.00587/) (2020)
- [A causal view of compositional zero-shot recognition](https://mlsanity.com/paper/2006.14610/) (2020)
- [Learning to Compose Soft Prompts for Compositional Zero-Shot Learning](https://mlsanity.com/paper/2204.03574/) (2022)
- [Disentangling Visual Embeddings for Attributes and Objects](https://mlsanity.com/paper/2205.08536/) (2022)
- [Siamese Contrastive Embedding Network for Compositional Zero-Shot Learning](https://mlsanity.com/paper/2206.14475/) (2022)
- [Learning Graph Embeddings for Open World Compositional Zero-Shot Learning](https://mlsanity.com/paper/2105.01017/) (2021)
- [Independent Prototype Propagation for Zero-Shot Compositionality](https://mlsanity.com/paper/2106.00305/) (2021)
- [KG-SP: Knowledge Guided Simple Primitives for Open World Compositional Zero-Shot Learning](https://mlsanity.com/paper/2205.06784/) (2022)
- [Joint Discovery of Object States and Manipulation Actions](https://mlsanity.com/paper/1702.02738/) (2017)
- [Learning Invariant Visual Representations for Compositional Zero-Shot Learning](https://mlsanity.com/paper/2206.00415/) (2022)

## Built on by (most cited first, 29 of 29)

- [Context-based and Diversity-driven Specificity in Compositional Zero-Shot Learning](https://mlsanity.com/paper/2402.17251/) (2024)
- [Learning Clustering-based Prototypes for Compositional Zero-shot Learning](https://mlsanity.com/paper/2502.06501/) (2025)
- [Compositional Zero-shot Learning via Progressive Language-based Observations](https://mlsanity.com/paper/2311.14749/) (2023)
- [C2C: Component-to-Composition Learning for Zero-Shot Compositional Action Recognition](https://mlsanity.com/paper/2407.06113/) (2024)
- [CusConcept: Customized Visual Concept Decomposition with Diffusion Models](https://mlsanity.com/paper/2410.00398/) (2024)
- [DRC: Enhancing Personalized Image Generation via Disentangled Representation Composition](https://mlsanity.com/paper/2504.17349/) (2025)
- [Dessie: Disentanglement for Articulated 3D Horse Shape and Pose Estimation from Images](https://mlsanity.com/paper/2410.03438/) (2024)
- [On the Element-Wise Representation and Reasoning in Zero-Shot Image Recognition: A Systematic Survey](https://mlsanity.com/paper/2408.04879/) (2024)
- [Enhancing Compositional Generalization via Compositional Feature Alignment](https://mlsanity.com/paper/2402.02851/) (2024)
- [HOPE: A Memory-Based and Composition-Aware Framework for Zero-Shot Learning with Hopfield Network and Soft Mixture of Experts](https://mlsanity.com/paper/2311.14747/) (2023)
- [Multi-Objective Instruction-Aware Representation Learning in Procedural Content Generation RL](https://mlsanity.com/paper/2508.09193/) (2025)
- [FlowComposer: Composable Flows for Compositional Zero-Shot Learning](https://mlsanity.com/paper/2603.16641/) (2026)
- [Not Just Object, But State: Compositional Incremental Learning without Forgetting](https://mlsanity.com/paper/2411.01739/) (2024)
- [Visual Adaptive Prompting for Compositional Zero-Shot Learning](https://mlsanity.com/paper/2502.20292/) (2025)
- [A Conditional Probability Framework for Compositional Zero-shot Learning](https://mlsanity.com/paper/2507.17377/) (2025)
- [CSCNET: Class-Specified Cascaded Network for Compositional Zero-Shot Learning](https://mlsanity.com/paper/2403.05924/) (2024)
- [Leveraging MLLM Embeddings and Attribute Smoothing for Compositional Zero-Shot Learning](https://mlsanity.com/paper/2411.12584/) (2024)
- [Unified Framework for Open-World Compositional Zero-shot Learning](https://mlsanity.com/paper/2412.04083/) (2024)
- [MSCI: Addressing CLIP's Inherent Limitations for Compositional Zero-Shot Learning](https://mlsanity.com/paper/2505.10289/) (2025)
- [CAMS: Towards Compositional Zero-Shot Learning via Gated Cross-Attention and Multi-Space Disentanglement](https://mlsanity.com/paper/2511.16378/) (2025)
- [H2em: Learning Hierarchical Hyperbolic Embeddings for Compositional Zero-Shot Learning](https://mlsanity.com/paper/2512.20029/) (2025)
- [SEER-ZSL: Semantic Encoder-Enhanced Representations for Generalized Zero-Shot Learning](https://mlsanity.com/paper/2312.13100/) (2023)
- [Compositional Zero-Shot Learning for Attribute-Based Object Reference in Human-Robot Interaction](https://mlsanity.com/paper/2312.13655/) (2023)
- [RevCD -- Reversed Conditional Diffusion for Generalized Zero-Shot Learning](https://mlsanity.com/paper/2409.00511/) (2024)
- [Semantically Guided Dynamic Visual Prototype Refinement for Compositional Zero-Shot Learning](https://mlsanity.com/paper/2501.07114/) (2025)
- [Proactive Disentangled Modeling of Trigger-Object Pairings for Backdoor Defense](https://mlsanity.com/paper/2508.01932/) (2025)
- [Prompt-Based Continual Compositional Zero-Shot Learning](https://mlsanity.com/paper/2512.09172/) (2025)
- [Structure-aware Prompt Adaptation from Seen to Unseen for Open-Vocabulary Compositional Zero-Shot Learning](https://mlsanity.com/paper/2603.03815/) (2026)
- [Progressive Reasoning with Primitive Correction for Compositional Zero-Shot Learning](https://mlsanity.com/paper/2607.05911/) (2026)

## Abstract (arXiv)

Compositional zero-shot learning (CZSL) aims at learning visual concepts (i.e., attributes and objects) from seen compositions and combining concept knowledge into unseen compositions. The key to CZSL is learning the disentanglement of the attribute-object composition. To this end, we propose to exploit cross-attentions as compositional disentanglers to learn disentangled concept embeddings. For example, if we want to recognize an unseen composition "yellow flower", we can learn the attribute concept "yellow" and object concept "flower" from different yellow objects and different flowers respectively. To further constrain the disentanglers to learn the concept of interest, we employ a regularization at the attention level. Specifically, we adapt the earth mover's distance (EMD) as a feature similarity metric in the cross-attention module. Moreover, benefiting from concept disentanglement, we improve the inference process and tune the prediction score by combining multiple concept probabilities. Comprehensive experiments on three CZSL benchmark datasets demonstrate that our method significantly outperforms previous works in both closed- and open-world settings, establishing a new state-of-the-art.