Continual Learning for Generative AI: From LLMs to MLLMs and Beyond

• URL: http://arxiv.org/abs/2506.13045v4
• Date: Sun, 24 Aug 2025 10:01:04 GMT
• Title: Continual Learning for Generative AI: From LLMs to MLLMs and Beyond
• Authors: Haiyang Guo, Fanhu Zeng, Fei Zhu, Jiayi Wang, Xukai Wang, Jingang Zhou, Hongbo Zhao, Wenzhuo Liu, Shijie Ma, Da-Han Wang, Xu-Yao Zhang, Cheng-Lin Liu,
• Abstract summary: We present a comprehensive survey of continual learning methods for mainstream generative AI models.<n>We categorize these approaches into three paradigms: architecture-based, regularization-based, and replay-based.<n>We analyze continual learning setups for different generative models, including training objectives, benchmarks, and core backbones.
• Score: 56.29231194002407
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The rapid advancement of generative models has empowered modern AI systems to comprehend and produce highly sophisticated content, even achieving human-level performance in specific domains. However, these models are fundamentally constrained by \emph{catastrophic forgetting}, \ie~a persistent challenge where models experience performance degradation on previously learned tasks when adapting to new tasks. To address this practical limitation, numerous approaches have been proposed to enhance the adaptability and scalability of generative AI in real-world applications. In this work, we present a comprehensive survey of continual learning methods for mainstream generative AI models, encompassing large language models, multimodal large language models, vision-language-action models, and diffusion models. Drawing inspiration from the memory mechanisms of the human brain, we systematically categorize these approaches into three paradigms: architecture-based, regularization-based, and replay-based methods, while elucidating their underlying methodologies and motivations. We further analyze continual learning setups for different generative models, including training objectives, benchmarks, and core backbones, thereby providing deeper insights into the field. The project page of this paper is available at https://github.com/Ghy0501/Awesome-Continual-Learning-in-Generative-Models.

Related papers

• Speed Always Wins: A Survey on Efficient Architectures for Large Language Models [51.817121227562964]
  Large Language Models (LLMs) have delivered impressive results in language understanding, generation, reasoning, and pushes the ability boundary of multimodal models.<n> Transformer models, as the foundation of modern LLMs, offer a strong baseline with excellent scaling properties.<n>The traditional transformer architecture requires substantial computations and poses significant obstacles for large-scale training and practical deployment.
  arXiv  Detail & Related papers  (2025-08-13T14:13:46Z)
• A Survey on Generative Model Unlearning: Fundamentals, Taxonomy, Evaluation, and Future Direction [21.966560704390716]
  We review current research on Generative Model Unlearning (GenMU)<n>We propose a unified analytical framework for categorizing unlearning objectives, methodological strategies, and evaluation metrics.<n>We highlight the potential practical value of unlearning techniques in real-world applications.
  arXiv  Detail & Related papers  (2025-07-26T09:49:57Z)
• From Task-Specific Models to Unified Systems: A Review of Model Merging Approaches [13.778158813149833]
  This paper establishes a new taxonomy of model merging methods, systematically comparing different approaches and providing an overview of key developments.<n>Despite the rapid progress in this field, a comprehensive taxonomy and survey summarizing recent advances and predicting future directions are still lacking.
  arXiv  Detail & Related papers  (2025-03-12T02:17:31Z)
• Few-Shot Class-Incremental Model Attribution Using Learnable Representation From CLIP-ViT Features [1.534667887016089]
  This work proposes a new strategy to deal with persistently emerging generative models.<n>We adapt few-shot class-incremental learning (FSCIL) mechanisms for MA problem to uncover novel generative AI models.<n>To learn an effective representation, we propose Adaptive Integration Module (AIM) to calculate a weighted sum of CLIP-ViT block features for each image.
  arXiv  Detail & Related papers  (2025-03-11T08:05:26Z)
• A Survey of Model Architectures in Information Retrieval [59.61734783818073]
  The period from 2019 to the present has represented one of the biggest paradigm shifts in information retrieval (IR) and natural language processing (NLP)<n>We trace the development from traditional term-based methods to modern neural approaches, particularly highlighting the impact of transformer-based models and subsequent large language models (LLMs)<n>We conclude with a forward-looking discussion of emerging challenges and future directions.
  arXiv  Detail & Related papers  (2025-02-20T18:42:58Z)
• Generating Computational Cognitive Models using Large Language Models [13.201934636532577]
  We introduce a pipeline for Guided generation of Computational Cognitive Models (GeCCo)<n>GeCCo prompts an LLM to propose candidate models, fits proposals to held-out data, and iteratively refines them based on their predictive performance.<n>We benchmark this approach across four different cognitive domains.
  arXiv  Detail & Related papers  (2025-02-02T19:07:13Z)
• Boosting Alignment for Post-Unlearning Text-to-Image Generative Models [55.82190434534429]
  Large-scale generative models have shown impressive image-generation capabilities, propelled by massive data.<n>This often inadvertently leads to the generation of harmful or inappropriate content and raises copyright concerns.<n>We propose a framework that seeks an optimal model update at each unlearning iteration, ensuring monotonic improvement on both objectives.
  arXiv  Detail & Related papers  (2024-12-09T21:36:10Z)
• Unified Generative and Discriminative Training for Multi-modal Large Language Models [88.84491005030316]
  Generative training has enabled Vision-Language Models (VLMs) to tackle various complex tasks. Discriminative training, exemplified by models like CLIP, excels in zero-shot image-text classification and retrieval. This paper proposes a unified approach that integrates the strengths of both paradigms.
  arXiv  Detail & Related papers  (2024-11-01T01:51:31Z)
• Explanation, Debate, Align: A Weak-to-Strong Framework for Language Model Generalization [0.6629765271909505]
  This paper introduces a novel approach to model alignment through weak-to-strong generalization in the context of language models. Our results suggest that this facilitation-based approach not only enhances model performance but also provides insights into the nature of model alignment.
  arXiv  Detail & Related papers  (2024-09-11T15:16:25Z)
• Model Merging in LLMs, MLLMs, and Beyond: Methods, Theories, Applications and Opportunities [89.40778301238642]
  Model merging is an efficient empowerment technique in the machine learning community. There is a significant gap in the literature regarding a systematic and thorough review of these techniques.
  arXiv  Detail & Related papers  (2024-08-14T16:58:48Z)
• Deep Generative Models in Robotics: A Survey on Learning from Multimodal Demonstrations [52.11801730860999]
  In recent years, the robot learning community has shown increasing interest in using deep generative models to capture the complexity of large datasets. We present the different types of models that the community has explored, such as energy-based models, diffusion models, action value maps, or generative adversarial networks. We also present the different types of applications in which deep generative models have been used, from grasp generation to trajectory generation or cost learning.
  arXiv  Detail & Related papers  (2024-08-08T11:34:31Z)
• Enhancing Generative Class Incremental Learning Performance with Model Forgetting Approach [50.36650300087987]
  This study presents a novel approach to Generative Class Incremental Learning (GCIL) by introducing the forgetting mechanism. We have found that integrating the forgetting mechanisms significantly enhances the models' performance in acquiring new knowledge.
  arXiv  Detail & Related papers  (2024-03-27T05:10:38Z)
• Foundation Models for Decision Making: Problems, Methods, and Opportunities [124.79381732197649]
  Foundation models pretrained on diverse data at scale have demonstrated extraordinary capabilities in a wide range of vision and language tasks. New paradigms are emerging for training foundation models to interact with other agents and perform long-term reasoning. Research at the intersection of foundation models and decision making holds tremendous promise for creating powerful new systems.
  arXiv  Detail & Related papers  (2023-03-07T18:44:07Z)
• Towards Interpretable Deep Reinforcement Learning Models via Inverse Reinforcement Learning [27.841725567976315]
  We propose a novel framework utilizing Adversarial Inverse Reinforcement Learning. This framework provides global explanations for decisions made by a Reinforcement Learning model. We capture intuitive tendencies that the model follows by summarizing the model's decision-making process.
  arXiv  Detail & Related papers  (2022-03-30T17:01:59Z)
• Deep Model-Based Reinforcement Learning for High-Dimensional Problems, a Survey [1.2031796234206134]
  Model-based reinforcement learning creates an explicit model of the environment dynamics to reduce the need for environment samples. A challenge for deep model-based methods is to achieve high predictive power while maintaining low sample complexity. We propose a taxonomy based on three approaches: using explicit planning on given transitions, using explicit planning on learned transitions, and end-to-end learning of both planning and transitions.
  arXiv  Detail & Related papers  (2020-08-11T08:49:04Z)
• Goal-Aware Prediction: Learning to Model What Matters [105.43098326577434]
  One of the fundamental challenges in using a learned forward dynamics model is the mismatch between the objective of the learned model and that of the downstream planner or policy. We propose to direct prediction towards task relevant information, enabling the model to be aware of the current task and encouraging it to only model relevant quantities of the state space. We find that our method more effectively models the relevant parts of the scene conditioned on the goal, and as a result outperforms standard task-agnostic dynamics models and model-free reinforcement learning.
  arXiv  Detail & Related papers  (2020-07-14T16:42:59Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.