# Analyzing and Reducing Catastrophic Forgetting in Parameter Efficient Tuning

arXiv: https://arxiv.org/abs/2402.18865 · HTML: https://mlsanity.com/paper/2402.18865/

## Key facts

- Cited by 60 papers in the map (data through 14 Oct 2026).
- Builds on 27 of its 36 references that are in the map.
- First posted on arXiv 29 Feb 2024.

## Builds on (most cited first, 27 of 27)

- [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)
- [PIQA: Reasoning about Physical Commonsense in Natural Language](https://mlsanity.com/paper/1911.11641/) (2019)
- [Learning without Forgetting](https://mlsanity.com/paper/1606.09282/) (2016)
- [Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering](https://mlsanity.com/paper/2209.09513/) (2022)
- [Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Them](https://mlsanity.com/paper/2210.09261/) (2022)
- [Experience Replay for Continual Learning](https://mlsanity.com/paper/1811.11682/) (2018)
- [Efficient Lifelong Learning with A-GEM](https://mlsanity.com/paper/1812.00420/) (2018)
- [Learning to Prompt for Continual Learning](https://mlsanity.com/paper/2112.08654/) (2021)
- [Dark Experience for General Continual Learning: a Strong, Simple Baseline](https://mlsanity.com/paper/2004.07211/) (2020)
- [Linear Mode Connectivity and the Lottery Ticket Hypothesis](https://mlsanity.com/paper/1912.05671/) (2019)
- [Loss Surfaces, Mode Connectivity, and Fast Ensembling of DNNs](https://mlsanity.com/paper/1802.10026/) (2018)
- [Gradient Projection Memory for Continual Learning](https://mlsanity.com/paper/2103.09762/) (2021)
- [The Role of Permutation Invariance in Linear Mode Connectivity of Neural Networks](https://mlsanity.com/paper/2110.06296/) (2021)
- [Progressive Prompts: Continual Learning for Language Models](https://mlsanity.com/paper/2301.12314/) (2023)
- [Linear Mode Connectivity in Multitask and Continual Learning](https://mlsanity.com/paper/2010.04495/) (2020)
- [Bridging Mode Connectivity in Loss Landscapes and Adversarial Robustness](https://mlsanity.com/paper/2005.00060/) (2020)
- [A Unified Continual Learning Framework with General Parameter-Efficient Tuning](https://mlsanity.com/paper/2303.10070/) (2023)
- [JEC-QA: A Legal-Domain Question Answering Dataset](https://mlsanity.com/paper/1911.12011/) (2019)
- [NumGLUE: A Suite of Fundamental yet Challenging Mathematical Reasoning Tasks](https://mlsanity.com/paper/2204.05660/) (2022)
- [Continual Learning via Neural Pruning](https://mlsanity.com/paper/1903.04476/) (2019)
- [Trillion Dollar Words: A New Financial Dataset, Task & Market Analysis](https://mlsanity.com/paper/2305.07972/) (2023)
- [Generative Models from the perspective of Continual Learning](https://mlsanity.com/paper/1812.09111/) (2018)
- [Introducing Language Guidance in Prompt-based Continual Learning](https://mlsanity.com/paper/2308.15827/) (2023)
- [Achieving a Better Stability-Plasticity Trade-off via Auxiliary Networks in Continual Learning](https://mlsanity.com/paper/2303.09483/) (2023)
- [Continual Learning Beyond a Single Model](https://mlsanity.com/paper/2202.09826/) (2022)

## Built on by (most cited first, 30 of 60)

- [A Survey on LoRA of Large Language Models](https://mlsanity.com/paper/2407.11046/) (2024)
- [Towards Lifelong Learning of Large Language Models: A Survey](https://mlsanity.com/paper/2406.06391/) (2024)
- [Image Conductor: Precision Control for Interactive Video Synthesis](https://mlsanity.com/paper/2406.15339/) (2024)
- [Tensor Product Attention Is All You Need](https://mlsanity.com/paper/2501.06425/) (2025)
- [Pre-Act: Multi-Step Planning and Reasoning Improves Acting in LLM Agents](https://mlsanity.com/paper/2505.09970/) (2025)
- [Recurrent Knowledge Identification and Fusion for Language Model Continual Learning](https://mlsanity.com/paper/2502.17510/) (2025)
- [MoFO: Momentum-Filtered Optimizer for Mitigating Forgetting in LLM Fine-Tuning](https://mlsanity.com/paper/2407.20999/) (2024)
- [Refining Input Guardrails: Enhancing LLM-as-a-Judge Efficiency Through Chain-of-Thought Fine-Tuning and Alignment](https://mlsanity.com/paper/2501.13080/) (2025)
- [How to Alleviate Catastrophic Forgetting in LLMs Finetuning? Hierarchical Layer-Wise and Element-Wise Regularization](https://mlsanity.com/paper/2501.13669/) (2025)
- [Human-inspired Perspectives: A Survey on AI Long-term Memory](https://mlsanity.com/paper/2411.00489/) (2024)
- [Continual Learning for Generative AI: From LLMs to MLLMs and Beyond](https://mlsanity.com/paper/2506.13045/) (2025)
- [Parameter-Efficient Continual Fine-Tuning: A Survey](https://mlsanity.com/paper/2504.13822/) (2025)
- [Emotion-Qwen: A Unified Framework for Emotion and Vision Understanding](https://mlsanity.com/paper/2505.06685/) (2025)
- [How to Teach Large Multimodal Models New Skills](https://mlsanity.com/paper/2510.08564/) (2025)
- [Too Helpful, Too Harmless, Too Honest or Just Right?](https://mlsanity.com/paper/2509.08486/) (2025)
- [Continual Learning in Vision-Language Models via Aligned Model Merging](https://mlsanity.com/paper/2506.03189/) (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)
- [On-Policy Replay for Continual Supervised Fine-Tuning](https://mlsanity.com/paper/2605.29495/) (2026)
- [KD-LoRA: A Hybrid Approach to Efficient Fine-Tuning with LoRA and Knowledge Distillation](https://mlsanity.com/paper/2410.20777/) (2024)
- [EF-LLM: Energy Forecasting LLM with AI-assisted Automation, Enhanced Sparse Prediction, Hallucination Detection](https://mlsanity.com/paper/2411.00852/) (2024)
- [Do Self-Evolving Agents Forget? Capability Degradation and Preservation in Lifelong LLM Agent Adaptation](https://mlsanity.com/paper/2605.09315/) (2026)
- [SPD-CFL: Stepwise Parameter Dropout for Efficient Continual Federated Learning](https://mlsanity.com/paper/2405.09394/) (2024)
- [Soft Language Prompts for Language Transfer](https://mlsanity.com/paper/2407.02317/) (2024)
- [PEFT-as-an-Attack! Jailbreaking Language Models during Federated Parameter-Efficient Fine-Tuning](https://mlsanity.com/paper/2411.19335/) (2024)
- [AIMMerging: Adaptive Iterative Model Merging Using Training Trajectories for Language Model Continual Learning](https://mlsanity.com/paper/2509.17348/) (2025)
- [How Much is Too Much? Exploring LoRA Rank Trade-offs for Retaining Knowledge and Domain Robustness](https://mlsanity.com/paper/2512.15634/) (2025)
- [Multi-source Unsupervised Domain Adaptation on Graphs with Transferability Modeling](https://mlsanity.com/paper/2406.10425/) (2024)
- [CALICO: Part-Focused Semantic Co-Segmentation with Large Vision-Language Models](https://mlsanity.com/paper/2412.19331/) (2024)
- [Enhancing Human-Like Responses in Large Language Models](https://mlsanity.com/paper/2501.05032/) (2025)

## Abstract (arXiv)

Existing research has shown that large language models (LLMs) exhibit remarkable performance in language understanding and generation. However, when LLMs are continuously fine-tuned on complex and diverse domain-specific downstream tasks, the inference performance on historical tasks decreases dramatically, which is known as a catastrophic forgetting problem. A trade-off needs to be kept between learning plasticity and memory stability. Plenty of existing works have explored strategies like memory replay, regularization and parameter isolation, but little is known about the geometric connection of various adjacent minima in the continual LLMs fine-tuning scenarios. In this work, we investigate the geometric connections of different minima through the lens of mode connectivity, which means different minima can be connected by a low-loss valley. Through extensive experiments, we uncover the mode connectivity phenomenon in the LLMs continual learning scenario and find that it can strike a balance between plasticity and stability. Building upon these findings, we propose a simple yet effective method called Interpolation-based LoRA (I-LoRA), which constructs a dual-memory experience replay framework based on LoRA parameter interpolations. Extensive experiments and analysis on eight domain-specific CL benchmarks demonstrate that I-LoRA consistently show significant improvement over the previous state-of-the-art approaches with up to 11% performance gains, providing a strong baseline and insights for future research on the large language model continual learning problem. Our code is available at urlhttps://github.com/which47/LLMCL.