# Comprehensive Generative Replay for Task-Incremental Segmentation with Concurrent Appearance and Semantic Forgetting

arXiv: https://arxiv.org/abs/2406.19796 · HTML: https://mlsanity.com/paper/2406.19796/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 28 references that are in the map.
- First posted on arXiv 28 Jun 2024.

## Builds on (most cited first, 21 of 21)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [Classifier-Free Diffusion Guidance](https://mlsanity.com/paper/2207.12598/) (2022)
- [Overcoming catastrophic forgetting in neural networks](https://mlsanity.com/paper/1612.00796/) (2016)
- [Learning without Forgetting](https://mlsanity.com/paper/1606.09282/) (2016)
- [CLIP-Driven Universal Model for Organ Segmentation and Tumor Detection](https://mlsanity.com/paper/2301.00785/) (2023)
- [One Transformer Fits All Distributions in Multi-Modal Diffusion at Scale](https://mlsanity.com/paper/2303.06555/) (2023)
- [PLOP: Learning without Forgetting for Continual Semantic Segmentation](https://mlsanity.com/paper/2011.11390/) (2020)
- [MS-Net: Multi-Site Network for Improving Prostate Segmentation with Heterogeneous MRI Data](https://mlsanity.com/paper/2002.03366/) (2020)
- [STU-Net: Scalable and Transferable Medical Image Segmentation Models Empowered by Large-Scale Supervised Pre-training](https://mlsanity.com/paper/2304.06716/) (2023)
- [DoFE: Domain-oriented Feature Embedding for Generalizable Fundus Image Segmentation on Unseen Datasets](https://mlsanity.com/paper/2010.06208/) (2020)
- [JointNet: Extending Text-to-Image Diffusion for Dense Distribution Modeling](https://mlsanity.com/paper/2310.06347/) (2023)
- [Learning Incrementally to Segment Multiple Organs in a CT Image](https://mlsanity.com/paper/2203.02100/) (2022)
- [Domain-incremental Cardiac Image Segmentation with Style-oriented Replay and Domain-sensitive Feature Whitening](https://mlsanity.com/paper/2211.04862/) (2022)
- [Inherit with Distillation and Evolve with Contrast: Exploring Class Incremental Semantic Segmentation Without Exemplar Memory](https://mlsanity.com/paper/2309.15413/) (2023)
- [Incremental Learning for Heterogeneous Structure Segmentation in Brain Tumor MRI](https://mlsanity.com/paper/2305.19404/) (2023)
- [Generative appearance replay for continual unsupervised domain adaptation](https://mlsanity.com/paper/2301.01211/) (2023)
- [Learning towards Synchronous Network Memorizability and Generalizability for Continual Segmentation across Multiple Sites](https://mlsanity.com/paper/2206.06813/) (2022)

## Built on by (most cited first, 3 of 3)

- [ReplayCAD: Generative Diffusion Replay for Continual Anomaly Detection](https://mlsanity.com/paper/2505.06603/) (2025)
- [F^2TTA: Free-Form Test-Time Adaptation on Cross-Domain Medical Image Classification via Image-Level Disentangled Prompt Tuning](https://mlsanity.com/paper/2507.02437/) (2025)
- [C^2GR: Coupled Comprehensive Generative Replay for a Continually Learnable Universal Segmentation Model](https://mlsanity.com/paper/2606.23473/) (2026)

## Abstract (arXiv)

Generalist segmentation models are increasingly favored for diverse tasks involving various objects from different image sources. Task-Incremental Learning (TIL) offers a privacy-preserving training paradigm using tasks arriving sequentially, instead of gathering them due to strict data sharing policies. However, the task evolution can span a wide scope that involves shifts in both image appearance and segmentation semantics with intricate correlation, causing concurrent appearance and semantic forgetting. To solve this issue, we propose a Comprehensive Generative Replay (CGR) framework that restores appearance and semantic knowledge by synthesizing image-mask pairs to mimic past task data, which focuses on two aspects: modeling image-mask correspondence and promoting scalability for diverse tasks. Specifically, we introduce a novel Bayesian Joint Diffusion (BJD) model for high-quality synthesis of image-mask pairs with their correspondence explicitly preserved by conditional denoising. Furthermore, we develop a Task-Oriented Adapter (TOA) that recalibrates prompt embeddings to modulate the diffusion model, making the data synthesis compatible with different tasks. Experiments on incremental tasks (cardiac, fundus and prostate segmentation) show its clear advantage for alleviating concurrent appearance and semantic forgetting. Code is available at https://github.com/jingyzhang/CGR.