# Compositional Few-Shot Class-Incremental Learning

arXiv: https://arxiv.org/abs/2405.17022 · HTML: https://mlsanity.com/paper/2405.17022/

## Key facts

- Cited by 18 papers in the map (data through 14 Oct 2026).
- Builds on 25 of its 40 references that are in the map.
- First posted on arXiv 27 May 2024.

## Builds on (most cited first, 25 of 25)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [Learning Deep Features for Discriminative Localization](https://mlsanity.com/paper/1512.04150/) (2015)
- [iCaRL: Incremental Classifier and Representation Learning](https://mlsanity.com/paper/1611.07725/) (2016)
- [Matching Networks for One Shot Learning](https://mlsanity.com/paper/1606.04080/) (2016)
- [Similarity of Neural Network Representations Revisited](https://mlsanity.com/paper/1905.00414/) (2019)
- [FitNets: Hints for Thin Deep Nets](https://mlsanity.com/paper/1412.6550/) (2014)
- [End-to-End Incremental Learning](https://mlsanity.com/paper/1807.09536/) (2018)
- [Few-Shot Class-Incremental Learning](https://mlsanity.com/paper/2004.10956/) (2020)
- [Few-Shot Incremental Learning with Continually Evolved Classifiers](https://mlsanity.com/paper/2104.03047/) (2021)
- [Forward Compatible Few-Shot Class-Incremental Learning](https://mlsanity.com/paper/2203.06953/) (2022)
- [Task-Driven Modular Networks for Zero-Shot Compositional Learning](https://mlsanity.com/paper/1905.05908/) (2019)
- [Few-Shot Classification with Feature Map Reconstruction Networks](https://mlsanity.com/paper/2012.01506/) (2020)
- [Neural Collapse Inspired Feature-Classifier Alignment for Few-Shot Class Incremental Learning](https://mlsanity.com/paper/2302.03004/) (2023)
- [Reliability of CKA as a Similarity Measure in Deep Learning](https://mlsanity.com/paper/2210.16156/) (2022)
- [Few-Shot Class-Incremental Learning from an Open-Set Perspective](https://mlsanity.com/paper/2208.00147/) (2022)
- [Learning with Fantasy: Semantic-Aware Virtual Contrastive Constraint for Few-Shot Class-Incremental Learning](https://mlsanity.com/paper/2304.00426/) (2023)
- [Margin-Based Few-Shot Class-Incremental Learning with Class-Level Overfitting Mitigation](https://mlsanity.com/paper/2210.04524/) (2022)
- [Understanding Cross-Domain Few-Shot Learning Based on Domain Similarity and Few-Shot Difficulty](https://mlsanity.com/paper/2202.01339/) (2022)
- [On the Soft-Subnetwork for Few-shot Class Incremental Learning](https://mlsanity.com/paper/2209.07529/) (2022)
- [Revisiting Pose-Normalization for Fine-Grained Few-Shot Recognition](https://mlsanity.com/paper/2004.00705/) (2020)
- [Baby steps towards few-shot learning with multiple semantics](https://mlsanity.com/paper/1906.01905/) (2019)
- [Matching Feature Sets for Few-Shot Image Classification](https://mlsanity.com/paper/2204.00949/) (2022)
- [Compositional Few-Shot Recognition with Primitive Discovery and Enhancing](https://mlsanity.com/paper/2005.06047/) (2020)

## Built on by (most cited first, 18 of 18)

- [The Devil is in Low-Level Features for Cross-Domain Few-Shot Segmentation](https://mlsanity.com/paper/2503.21150/) (2025)
- [Revisiting Pool-based Prompt Learning for Few-shot Class-incremental Learning](https://mlsanity.com/paper/2507.09183/) (2025)
- [Mamba-FSCIL: Dynamic Adaptation with Selective State Space Model for Few-Shot Class-Incremental Learning](https://mlsanity.com/paper/2407.06136/) (2024)
- [Adapter Naturally Serves as Decoupler for Cross-Domain Few-Shot Semantic Segmentation](https://mlsanity.com/paper/2506.07376/) (2025)
- [Resource-Constrained Federated Continual Learning: What Does Matter?](https://mlsanity.com/paper/2501.08737/) (2025)
- [Latest Advancements Towards Catastrophic Forgetting under Data Scarcity: A Comprehensive Survey on Few-Shot Class Incremental Learning](https://mlsanity.com/paper/2502.08181/) (2025)
- [Random Registers for Cross-Domain Few-Shot Learning](https://mlsanity.com/paper/2506.02843/) (2025)
- [Learn by Reasoning: Analogical Weight Generation for Few-Shot Class-Incremental Learning](https://mlsanity.com/paper/2503.21258/) (2025)
- [Revisiting Continuity of Image Tokens for Cross-domain Few-shot Learning](https://mlsanity.com/paper/2506.03110/) (2025)
- [Remedying Target-Domain Astigmatism for Cross-Domain Few-Shot Object Detection](https://mlsanity.com/paper/2603.18541/) (2026)
- [Addressing Exacerbated Attention Sink for Source-Free Cross-Domain Few-Shot Learning](https://mlsanity.com/paper/2605.25799/) (2026)
- [Continuous Knowledge-Preserving Decomposition with Adaptive Layer Selection for Few-Shot Class-Incremental Learning](https://mlsanity.com/paper/2501.05017/) (2025)
- [Beyond Synthetic Replays: Turning Diffusion Features into Few-Shot Class-Incremental Learning Knowledge](https://mlsanity.com/paper/2503.23402/) (2025)
- [Adaptive Additive Parameter Updates of Vision Transformers for Few-Shot Continual Learning](https://mlsanity.com/paper/2504.08982/) (2025)
- [MoTiC: Momentum Tightness and Contrast for Few-Shot Class-Incremental Learning](https://mlsanity.com/paper/2509.19664/) (2025)
- [When Sensing Varies with Contexts: Context Probing for Tactile Few-Shot Class-Incremental Learning](https://mlsanity.com/paper/2603.25115/) (2026)
- [Unlocking Compositional Generalization in Continual Few-Shot Learning](https://mlsanity.com/paper/2605.11710/) (2026)
- [Free-Lunch Augmentation by Revisiting Diffusion-Based Data Generation for Cross-Domain Few-Shot Object Detection](https://mlsanity.com/paper/2608.04394/) (2026)

## Abstract (arXiv)

Few-shot class-incremental learning (FSCIL) is proposed to continually learn from novel classes with only a few samples after the (pre-)training on base classes with sufficient data. However, this remains a challenge. In contrast, humans can easily recognize novel classes with a few samples. Cognitive science demonstrates that an important component of such human capability is compositional learning. This involves identifying visual primitives from learned knowledge and then composing new concepts using these transferred primitives, making incremental learning both effective and interpretable. To imitate human compositional learning, we propose a cognitive-inspired method for the FSCIL task. We define and build a compositional model based on set similarities, and then equip it with a primitive composition module and a primitive reuse module. In the primitive composition module, we propose to utilize the Centered Kernel Alignment (CKA) similarity to approximate the similarity between primitive sets, allowing the training and evaluation based on primitive compositions. In the primitive reuse module, we enhance primitive reusability by classifying inputs based on primitives replaced with the closest primitives from other classes. Experiments on three datasets validate our method, showing it outperforms current state-of-the-art methods with improved interpretability. Our code is available at https://github.com/Zoilsen/Comp-FSCIL.