# Mitigating Catastrophic Forgetting in Large Language Models with Self-Synthesized Rehearsal

arXiv: https://arxiv.org/abs/2403.01244 · HTML: https://mlsanity.com/paper/2403.01244/

## Key facts

- Cited by 118 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 31 references that are in the map.
- First posted on arXiv 2 Mar 2024.

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

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Overcoming catastrophic forgetting in neural networks](https://mlsanity.com/paper/1612.00796/) (2016)
- [SimCSE: Simple Contrastive Learning of Sentence Embeddings](https://mlsanity.com/paper/2104.08821/) (2021)
- [Gradient Episodic Memory for Continual Learning](https://mlsanity.com/paper/1706.08840/) (2017)
- [Experience Replay for Continual Learning](https://mlsanity.com/paper/1811.11682/) (2018)
- [Super-NaturalInstructions: Generalization via Declarative Instructions on 1600+ NLP Tasks](https://mlsanity.com/paper/2204.07705/) (2022)
- [An Empirical Study of Catastrophic Forgetting in Large Language Models During Continual Fine-tuning](https://mlsanity.com/paper/2308.08747/) (2023)
- [Dreaming to Distill: Data-free Knowledge Transfer via DeepInversion](https://mlsanity.com/paper/1912.08795/) (2019)
- [Progressive Prompts: Continual Learning for Language Models](https://mlsanity.com/paper/2301.12314/) (2023)
- [Episodic Memory in Lifelong Language Learning](https://mlsanity.com/paper/1906.01076/) (2019)
- [Reinforced Continual Learning](https://mlsanity.com/paper/1805.12369/) (2018)
- [Fine-tuned Language Models are Continual Learners](https://mlsanity.com/paper/2205.12393/) (2022)
- [Always Be Dreaming: A New Approach for Data-Free Class-Incremental Learning](https://mlsanity.com/paper/2106.09701/) (2021)
- [Continual Learning for Text Classification with Information Disentanglement Based Regularization](https://mlsanity.com/paper/2104.05489/) (2021)
- [CPR: Classifier-Projection Regularization for Continual Learning](https://mlsanity.com/paper/2006.07326/) (2020)
- [Two-stage LLM Fine-tuning with Less Specialization and More Generalization](https://mlsanity.com/paper/2211.00635/) (2022)
- [CITB: A Benchmark for Continual Instruction Tuning](https://mlsanity.com/paper/2310.14510/) (2023)
- [Task Agnostic Representation Consolidation: a Self-supervised based Continual Learning Approach](https://mlsanity.com/paper/2207.06267/) (2022)
- [Neural Architecture Search for Class-incremental Learning](https://mlsanity.com/paper/1909.06686/) (2019)
- [Response Enhanced Semi-supervised Dialogue Query Generation](https://mlsanity.com/paper/2312.12713/) (2023)

## Built on by (most cited first, 30 of 118)

- [Continual Learning of Large Language Models: A Comprehensive Survey](https://mlsanity.com/paper/2404.16789/) (2024)
- [From RAG to Memory: Non-Parametric Continual Learning for Large Language Models](https://mlsanity.com/paper/2502.14802/) (2025)
- [The Emerged Security and Privacy of LLM Agent: A Survey with Case Studies](https://mlsanity.com/paper/2407.19354/) (2024)
- [A Survey on Self-Evolution of Large Language Models](https://mlsanity.com/paper/2404.14387/) (2024)
- [Towards Lifelong Learning of Large Language Models: A Survey](https://mlsanity.com/paper/2406.06391/) (2024)
- [Mitigating Catastrophic Forgetting in Language Transfer via Model Merging](https://mlsanity.com/paper/2407.08699/) (2024)
- [Self-MoE: Towards Compositional Large Language Models with Self-Specialized Experts](https://mlsanity.com/paper/2406.12034/) (2024)
- [R1-T1: Fully Incentivizing Translation Capability in LLMs via Reasoning Learning](https://mlsanity.com/paper/2502.19735/) (2025)
- [Unlocking the Power of Function Vectors for Characterizing and Mitigating Catastrophic Forgetting in Continual Instruction Tuning](https://mlsanity.com/paper/2502.11019/) (2025)
- [Recurrent Knowledge Identification and Fusion for Language Model Continual Learning](https://mlsanity.com/paper/2502.17510/) (2025)
- [Counteraction-Aware Multi-Teacher On-Policy Distillation for General Capability Recovery with Domain Preservation](https://mlsanity.com/paper/2605.27115/) (2026)
- [MoFO: Momentum-Filtered Optimizer for Mitigating Forgetting in LLM Fine-Tuning](https://mlsanity.com/paper/2407.20999/) (2024)
- [How Much Can We Forget about Data Contamination?](https://mlsanity.com/paper/2410.03249/) (2024)
- [Mitigating Forgetting in LLM Fine-Tuning via Low-Perplexity Token Learning](https://mlsanity.com/paper/2501.14315/) (2025)
- [Scalable Fingerprinting of Large Language Models](https://mlsanity.com/paper/2502.07760/) (2025)
- [Mitigating Catastrophic Forgetting in Large Language Models with Forgetting-aware Pruning](https://mlsanity.com/paper/2509.08255/) (2025)
- [Continual Learning for Generative AI: From LLMs to MLLMs and Beyond](https://mlsanity.com/paper/2506.13045/) (2025)
- [Towards Objective Fine-tuning: How LLMs' Prior Knowledge Causes Potential Poor Calibration?](https://mlsanity.com/paper/2505.20903/) (2025)
- [Continual Gradient Low-Rank Projection Fine-Tuning for LLMs](https://mlsanity.com/paper/2507.02503/) (2025)
- [How to Teach Large Multimodal Models New Skills](https://mlsanity.com/paper/2510.08564/) (2025)
- [Mapping Post-Training Forgetting in Language Models at Scale](https://mlsanity.com/paper/2510.17776/) (2025)
- [KIF: Knowledge Identification and Fusion for Language Model Continual Learning](https://mlsanity.com/paper/2408.05200/) (2024)
- [Systematic Knowledge Injection into Large Language Models via Diverse Augmentation for Domain-Specific RAG](https://mlsanity.com/paper/2502.08356/) (2025)
- [When Continue Learning Meets Multimodal Large Language Model: A Survey](https://mlsanity.com/paper/2503.01887/) (2025)
- [SelfAug: Mitigating Catastrophic Forgetting in Retrieval-Augmented Generation via Distribution Self-Alignment](https://mlsanity.com/paper/2509.03934/) (2025)
- [Demystifying Language Model Forgetting with Low-rank Example Associations](https://mlsanity.com/paper/2406.14026/) (2024)
- [Mitigating Memorization in LLMs using Activation Steering](https://mlsanity.com/paper/2503.06040/) (2025)
- [Improved Supervised Fine-Tuning for Large Language Models to Mitigate Catastrophic Forgetting](https://mlsanity.com/paper/2506.09428/) (2025)
- [FOREVER: Forgetting Curve-Inspired Memory Replay for Language Model Continual Learning](https://mlsanity.com/paper/2601.03938/) (2026)
- [A Scalable Measure of Loss Landscape Curvature for Analyzing the Training Dynamics of LLMs](https://mlsanity.com/paper/2601.16979/) (2026)

## Abstract (arXiv)

Large language models (LLMs) suffer from catastrophic forgetting during continual learning. Conventional rehearsal-based methods rely on previous training data to retain the model's ability, which may not be feasible in real-world applications. When conducting continual learning based on a publicly-released LLM checkpoint, the availability of the original training data may be non-existent. To address this challenge, we propose a framework called Self-Synthesized Rehearsal (SSR) that uses the LLM to generate synthetic instances for rehearsal. Concretely, we first employ the base LLM for in-context learning to generate synthetic instances. Subsequently, we utilize the latest LLM to refine the instance outputs based on the synthetic inputs, preserving its acquired ability. Finally, we select diverse high-quality synthetic instances for rehearsal in future stages. Experimental results demonstrate that SSR achieves superior or comparable performance compared to conventional rehearsal-based approaches while being more data-efficient. Besides, SSR effectively preserves the generalization capabilities of LLMs in general domains.