# Feature-Space Generative Models for One-Shot Class-Incremental Learning

arXiv: https://arxiv.org/abs/2601.17905 · HTML: https://mlsanity.com/paper/2601.17905/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 32 of its 41 references that are in the map.
- First posted on arXiv 25 Jan 2026.

## Builds on (most cited first, 30 of 32)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Auto-Encoding Variational Bayes](https://mlsanity.com/paper/1312.6114/) (2013)
- [Prototypical Networks for Few-shot Learning](https://mlsanity.com/paper/1703.05175/) (2017)
- [Learning without Forgetting](https://mlsanity.com/paper/1606.09282/) (2016)
- [Continual Learning with Deep Generative Replay](https://mlsanity.com/paper/1705.08690/) (2017)
- [Learning to Compare: Relation Network for Few-Shot Learning](https://mlsanity.com/paper/1711.06025/) (2017)
- [Do Better ImageNet Models Transfer Better?](https://mlsanity.com/paper/1805.08974/) (2018)
- [Your Diffusion Model is Secretly a Zero-Shot Classifier](https://mlsanity.com/paper/2303.16203/) (2023)
- [CORe50: a New Dataset and Benchmark for Continuous Object Recognition](https://mlsanity.com/paper/1705.03550/) (2017)
- [Few-Shot Class-Incremental Learning](https://mlsanity.com/paper/2004.10956/) (2020)
- [Episodic Training for Domain Generalization](https://mlsanity.com/paper/1902.00113/) (2019)
- [SimpleShot: Revisiting Nearest-Neighbor Classification for Few-Shot Learning](https://mlsanity.com/paper/1911.04623/) (2019)
- [Few-Shot Incremental Learning with Continually Evolved Classifiers](https://mlsanity.com/paper/2104.03047/) (2021)
- [Free Lunch for Few-shot Learning: Distribution Calibration](https://mlsanity.com/paper/2101.06395/) (2021)
- [Visual Interpretability for Deep Learning: a Survey](https://mlsanity.com/paper/1802.00614/) (2018)
- [Incremental Learning Techniques for Semantic Segmentation](https://mlsanity.com/paper/1907.13372/) (2019)
- [Neural Collapse Inspired Feature-Classifier Alignment for Few-Shot Class Incremental Learning](https://mlsanity.com/paper/2302.03004/) (2023)
- [Semantic-aware Knowledge Distillation for Few-Shot Class-Incremental Learning](https://mlsanity.com/paper/2103.04059/) (2021)
- [Lifelong Machine Learning with Deep Streaming Linear Discriminant Analysis](https://mlsanity.com/paper/1909.01520/) (2019)
- [Delta-encoder: an effective sample synthesis method for few-shot object recognition](https://mlsanity.com/paper/1806.04734/) (2018)
- [Few-Shot Class-Incremental Learning by Sampling Multi-Phase Tasks](https://mlsanity.com/paper/2203.17030/) (2022)
- [Zero-Shot Learning via Semantic Similarity Embedding](https://mlsanity.com/paper/1509.04767/) (2015)
- [High Fidelity Visualization of What Your Self-Supervised Representation Knows About](https://mlsanity.com/paper/2112.09164/) (2021)
- [Few-Shot Lifelong Learning](https://mlsanity.com/paper/2103.00991/) (2021)
- [Class-Incremental Learning with Generative Classifiers](https://mlsanity.com/paper/2104.10093/) (2021)
- [Pre-trained Vision and Language Transformers Are Few-Shot Incremental Learners](https://mlsanity.com/paper/2404.02117/) (2024)
- [ProtoDiff: Learning to Learn Prototypical Networks by Task-Guided Diffusion](https://mlsanity.com/paper/2306.14770/) (2023)
- [iCub World: Friendly Robots Help Building Good Vision Data-Sets](https://mlsanity.com/paper/1306.3560/) (2013)

## Abstract (arXiv)

Few-shot class-incremental learning (FSCIL) is a paradigm where a model, initially trained on a dataset of base classes, must adapt to an expanding problem space by recognizing novel classes with limited data. We focus on the challenging FSCIL setup where a model receives only a single sample (1-shot) for each novel class and no further training or model alterations are allowed after the base training phase. This makes generalization to novel classes particularly difficult. We propose a novel approach predicated on the hypothesis that base and novel class embeddings have structural similarity. We map the original embedding space into a residual space by subtracting the class prototype (i.e., the average class embedding) of input samples. Then, we leverage generative modeling with VAE or diffusion models to learn the multi-modal distribution of residuals over the base classes, and we use this as a valuable structural prior to improve recognition of novel classes. Our approach, Gen1S, consistently improves novel class recognition over the state of the art across multiple benchmarks and backbone architectures.