# MOS: Model Surgery for Pre-Trained Model-Based Class-Incremental Learning

arXiv: https://arxiv.org/abs/2412.09441 · HTML: https://mlsanity.com/paper/2412.09441/

## Key facts

- Cited by 36 papers in the map (data through 14 Oct 2026).
- Builds on 27 of its 53 references that are in the map.
- First posted on arXiv 12 Dec 2024.

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

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Overcoming catastrophic forgetting in neural networks](https://mlsanity.com/paper/1612.00796/) (2016)
- [Prototypical Networks for Few-shot Learning](https://mlsanity.com/paper/1703.05175/) (2017)
- [Learning without Forgetting](https://mlsanity.com/paper/1606.09282/) (2016)
- [iCaRL: Incremental Classifier and Representation Learning](https://mlsanity.com/paper/1611.07725/) (2016)
- [Natural Adversarial Examples](https://mlsanity.com/paper/1907.07174/) (2019)
- [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)
- [AdaptFormer: Adapting Vision Transformers for Scalable Visual Recognition](https://mlsanity.com/paper/2205.13535/) (2022)
- [Learning multiple visual domains with residual adapters](https://mlsanity.com/paper/1705.08045/) (2017)
- [A Large-scale Study of Representation Learning with the Visual Task Adaptation Benchmark](https://mlsanity.com/paper/1910.04867/) (2019)
- [Scaling & Shifting Your Features: A New Baseline for Efficient Model Tuning](https://mlsanity.com/paper/2210.08823/) (2022)
- [Maintaining Discrimination and Fairness in Class Incremental Learning](https://mlsanity.com/paper/1911.07053/) (2019)
- [Semantic Drift Compensation for Class-Incremental Learning](https://mlsanity.com/paper/2004.00440/) (2020)
- [Mnemonics Training: Multi-Class Incremental Learning without Forgetting](https://mlsanity.com/paper/2002.10211/) (2020)
- [RanPAC: Random Projections and Pre-trained Models for Continual Learning](https://mlsanity.com/paper/2307.02251/) (2023)
- [Continual Learning with Pre-Trained Models: A Survey](https://mlsanity.com/paper/2401.16386/) (2024)
- [PILOT: A Pre-Trained Model-Based Continual Learning Toolbox](https://mlsanity.com/paper/2309.07117/) (2023)
- [Memory Efficient Class-Incremental Learning for Image Classification](https://mlsanity.com/paper/2008.01411/) (2020)
- [Benchmarking Omni-Vision Representation through the Lens of Visual Realms](https://mlsanity.com/paper/2207.07106/) (2022)
- [Visual Prompt Tuning in Null Space for Continual Learning](https://mlsanity.com/paper/2406.05658/) (2024)
- [Domain Prompt Learning with Quaternion Networks](https://mlsanity.com/paper/2312.08878/) (2023)
- [Revisiting Meta-Learning as Supervised Learning](https://mlsanity.com/paper/2002.00573/) (2020)
- [Generalized Knowledge Distillation via Relationship Matching](https://mlsanity.com/paper/2205.01915/) (2022)

## Built on by (most cited first, 30 of 36)

- [Integrating Task-Specific and Universal Adapters for Pre-Trained Model-based Class-Incremental Learning](https://mlsanity.com/paper/2508.08165/) (2025)
- [Mixture of Noise for Pre-Trained Model-Based Class-Incremental Learning](https://mlsanity.com/paper/2509.16738/) (2025)
- [External Knowledge Injection for CLIP-Based Class-Incremental Learning](https://mlsanity.com/paper/2503.08510/) (2025)
- [Fly-CL: A Fly-Inspired Framework for Enhancing Efficient Decorrelation and Reduced Training Time in Pre-trained Model-based Continual Representation Learning](https://mlsanity.com/paper/2510.16877/) (2025)
- [Navigating Semantic Drift in Task-Agnostic Class-Incremental Learning](https://mlsanity.com/paper/2502.07560/) (2025)
- [AnaCP: Toward Upper-Bound Continual Learning via Analytic Contrastive Projection](https://mlsanity.com/paper/2511.13880/) (2025)
- [CKAA: Cross-subspace Knowledge Alignment and Aggregation for Robust Continual Learning](https://mlsanity.com/paper/2507.09471/) (2025)
- [Enhancing Pretrained Model-based Continual Representation Learning via Guided Random Projection](https://mlsanity.com/paper/2603.19145/) (2026)
- [IncreFA: Breaking the Static Wall of Generative Model Attribution](https://mlsanity.com/paper/2604.17736/) (2026)
- [Energy-Structured Low-Rank Adaptation for Continual Learning](https://mlsanity.com/paper/2605.27482/) (2026)
- [C^2Prompt: Class-aware Client Knowledge Interaction for Federated Continual Learning](https://mlsanity.com/paper/2509.19674/) (2025)
- [The Lie of the Average: How Class Incremental Learning Evaluation Deceives You?](https://mlsanity.com/paper/2509.22580/) (2025)
- [Overcoming Catastrophic Forgetting in Visual Continual Learning with Reinforcement Fine-Tuning](https://mlsanity.com/paper/2605.09640/) (2026)
- [Boosting the Class-Incremental Learning in 3D Point Clouds via Zero-Collection-Cost Basic Shape Pre-Training](https://mlsanity.com/paper/2504.08412/) (2025)
- [PROL : Rehearsal Free Continual Learning in Streaming Data via Prompt Online Learning](https://mlsanity.com/paper/2507.12305/) (2025)
- [Temporal Imbalance of Positive and Negative Supervision in Class-Incremental Learning](https://mlsanity.com/paper/2603.02280/) (2026)
- [One Adapter for All: Towards Unified Representation in Step-Imbalanced Class-Incremental Learning](https://mlsanity.com/paper/2603.10237/) (2026)
- [HyCal: A Training-Free Prototype Calibration Method for Cross-Discipline Few-Shot Class-Incremental Learning](https://mlsanity.com/paper/2604.15678/) (2026)
- [Stable Routing for Mixture-of-Experts in Class-Incremental Learning](https://mlsanity.com/paper/2605.17571/) (2026)
- [Unlocking [CLS] Features for Continual Post-Training](https://mlsanity.com/paper/2502.14762/) (2025)
- [Partitioned Memory Storage Inspired Few-Shot Class-Incremental learning](https://mlsanity.com/paper/2504.20797/) (2025)
- [Adapt before Continual Learning](https://mlsanity.com/paper/2506.03956/) (2025)
- [MAFM^3: Modular Adaptation of Foundation Models for Multi-Modal Medical AI](https://mlsanity.com/paper/2511.11212/) (2025)
- [Pushing the Limits of Distillation-Based Continual Learning via Classifier-Proximal Lightweight Plugins](https://mlsanity.com/paper/2512.03537/) (2025)
- [From Isolation to Integration: Building an Adaptive Expert Forest for Pre-Trained Model-based Class-Incremental Learning](https://mlsanity.com/paper/2602.20911/) (2026)
- [LCA: Local Classifier Alignment for Continual Learning](https://mlsanity.com/paper/2603.09888/) (2026)
- [Pruned Adaptation Modules: A Simple yet Strong Baseline for Continual Foundation Models](https://mlsanity.com/paper/2603.21170/) (2026)
- [Bi-CRCL: Bidirectional Conservative-Radical Complementary Learning with Pre-trained Foundation Models for Class-incremental Medical Image Analysis](https://mlsanity.com/paper/2603.23729/) (2026)
- [Dual-Imbalance Continual Learning for Real-World Food Recognition](https://mlsanity.com/paper/2603.29133/) (2026)
- [Autonomous Drift Learning in Data Streams: A Unified Perspective](https://mlsanity.com/paper/2605.01295/) (2026)

## Abstract (arXiv)

Class-Incremental Learning (CIL) requires models to continually acquire knowledge of new classes without forgetting old ones. Despite Pre-trained Models (PTMs) have shown excellent performance in CIL, catastrophic forgetting still occurs as the model learns new concepts. Existing work seeks to utilize lightweight components to adjust the PTM, while the forgetting phenomenon still comes from em parameter and retrieval levels. Specifically, iterative updates of the model result in parameter drift, while mistakenly retrieving irrelevant modules leads to the mismatch during inference. To this end, we propose MOdel Surgery (MOS) to rescue the model from forgetting previous knowledge. By training task-specific adapters, we continually adjust the PTM to downstream tasks. To mitigate parameter-level forgetting, we present an adapter merging approach to learn task-specific adapters, which aims to bridge the gap between different components while reserve task-specific information. Besides, to address retrieval-level forgetting, we introduce a training-free self-refined adapter retrieval mechanism during inference, which leverages the model's inherent ability for better adapter retrieval. By jointly rectifying the model with those steps, MOS can robustly resist catastrophic forgetting in the learning process. Extensive experiments on seven benchmark datasets validate MOS's state-of-the-art performance. Code is available at: https://github.com/sun-hailong/AAAI25-MOS