# Task-Aware Feature Generation for Zero-Shot Compositional Learning

arXiv: https://arxiv.org/abs/1906.04854 · HTML: https://mlsanity.com/paper/1906.04854/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 11 Jun 2019.

## Built on by (most cited first, 11 of 11)

- [Learning Graph Embeddings for Compositional Zero-shot Learning](https://mlsanity.com/paper/2102.01987/) (2021)
- [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)
- [DRPT: Disentangled and Recurrent Prompt Tuning for Compositional Zero-Shot Learning](https://mlsanity.com/paper/2305.01239/) (2023)
- [Chop & Learn: Recognizing and Generating Object-State Compositions](https://mlsanity.com/paper/2309.14339/) (2023)
- [Distilled Reverse Attention Network for Open-world Compositional Zero-Shot Learning](https://mlsanity.com/paper/2303.00404/) (2023)
- [Revealing the Proximate Long-Tail Distribution in Compositional Zero-Shot Learning](https://mlsanity.com/paper/2312.15923/) (2023)
- [Generating Compositional Color Representations from Text](https://mlsanity.com/paper/2109.10477/) (2021)
- [Mutual Balancing in State-Object Components for Compositional Zero-Shot Learning](https://mlsanity.com/paper/2211.10647/) (2022)
- [Towards Zero-Shot Functional Compositionality of Language Models](https://mlsanity.com/paper/2303.03103/) (2023)
- [Attention Based Simple Primitives for Open World Compositional Zero-Shot Learning](https://mlsanity.com/paper/2407.13715/) (2024)

## Abstract (arXiv)

Visual concepts (e.g., red apple, big elephant) are often semantically compositional and each element of the compositions can be reused to construct novel concepts (e.g., red elephant). Compositional feature synthesis, which generates image feature distributions exploiting the semantic compositionality, is a promising approach to sample-efficient model generalization. In this work, we propose a task-aware feature generation (TFG) framework for compositional learning, which generates features of novel visual concepts by transferring knowledge from previously seen concepts. These synthetic features are then used to train a classifier to recognize novel concepts in a zero-shot manner. Our novel TFG design injects task-conditioned noise layer-by-layer, producing task-relevant variation at each level. We find the proposed generator design improves classification accuracy and sample efficiency. Our model establishes a new state of the art on three zero-shot compositional learning (ZSCL) benchmarks, outperforming the previous discriminative models by a large margin. Our model improves the performance of the prior arts by over 2x in the generalized ZSCL setting.