English
Related papers

Related papers: Soup to go: mitigating forgetting during continual…

200 papers

Multimodal models like LLaVA-1.5 achieve state-of-the-art visual understanding through visual instruction tuning on multitask datasets, enabling strong instruction-following and multimodal performance. However, multitask learning faces…

Computer Vision and Pattern Recognition · Computer Science 2025-06-16 Wenzhuo Liu , Fei Zhu , Haiyang Guo , Longhui Wei , Cheng-Lin Liu

Presently the most successful approaches to semi-supervised learning are based on consistency regularization, whereby a model is trained to be robust to small perturbations of its inputs and parameters. To understand consistency…

Machine Learning · Computer Science 2019-02-22 Ben Athiwaratkun , Marc Finzi , Pavel Izmailov , Andrew Gordon Wilson

End-to-end training of Spoken Language Models (SLMs) commonly involves adapting pre-trained text-based Large Language Models (LLMs) to the speech modality through multi-stage training on diverse tasks such as ASR, TTS and spoken question…

Computation and Language · Computer Science 2025-05-26 Chi-Yuan Hsiao , Ke-Han Lu , Kai-Wei Chang , Chih-Kai Yang , Wei-Chih Chen , Hung-yi Lee

In lifelong learning, data are used to improve performance not only on the present task, but also on past and future (unencountered) tasks. While typical transfer learning algorithms can improve performance on future tasks, their…

Learning a set of tasks over time, also known as continual learning (CL), is one of the most challenging problems in artificial intelligence due to catastrophic forgetting. Large language models (LLMs) are often impractical to frequent…

Machine Learning · Computer Science 2025-10-28 Jaya Krishna Mandivarapu

Model merging aims to build a multi-task learner by combining the parameters of individually fine-tuned models without additional training. While a straightforward approach is to average model parameters across tasks, this often results in…

Machine Learning · Computer Science 2025-04-04 Jiho Choi , Donggyun Kim , Chanhyuk Lee , Seunghoon Hong

Incrementally fine-tuning foundational models on new tasks or domains is now the de facto approach in NLP. A known pitfall of this approach is the \emph{catastrophic forgetting} of prior knowledge that happens during fine-tuning. A common…

Machine Learning · Computer Science 2025-02-12 Andrew Bai , Chih-Kuan Yeh , Cho-Jui Hsieh , Ankur Taly

Adapting language models (LMs) to new tasks via post-training carries the risk of degrading existing capabilities -- a phenomenon classically known as catastrophic forgetting. In this paper, toward identifying guidelines for mitigating this…

Machine Learning · Computer Science 2025-12-04 Howard Chen , Noam Razin , Karthik Narasimhan , Danqi Chen

Federated Learning (FL) has gained significant attraction due to its ability to enable privacy-preserving training over decentralized data. Current literature in FL mostly focuses on single-task learning. However, over time, new tasks may…

Machine Learning · Computer Science 2023-10-18 Yavuz Faruk Bakman , Duygu Nur Yaldiz , Yahya H. Ezzeldin , Salman Avestimehr

Continual learning is a promising machine learning paradigm to learn new tasks while retaining previously learned knowledge over streaming training data. Till now, rehearsal-based methods, keeping a small part of data from old tasks as a…

Machine Learning · Computer Science 2023-08-04 Quanziang Wang , Renzhen Wang , Yuexiang Li , Dong Wei , Kai Ma , Yefeng Zheng , Deyu Meng

Continual learning aims to acquire tasks sequentially without catastrophic forgetting, yet standard strategies face a core tradeoff: regularization-based methods (e.g., EWC) can overconstrain updates when task optima are weakly overlapping,…

Machine Learning · Computer Science 2026-05-28 Zekun Wang , Anant Gupta , Christopher J. MacLellan

Fine-tuning is a crucial paradigm for adapting pre-trained large language models to downstream tasks. Recently, methods like Low-Rank Adaptation (LoRA) have been shown to effectively fine-tune LLMs with an extreme reduction in trainable…

Machine Learning · Computer Science 2025-10-23 Reece Shuttleworth , Jacob Andreas , Antonio Torralba , Pratyusha Sharma

Lifelong learning requires models that can continuously learn from sequential streams of data without suffering catastrophic forgetting due to shifts in data distributions. Deep learning models have thrived in the non-sequential learning…

Computation and Language · Computer Science 2021-07-27 Nithin Holla , Pushkar Mishra , Helen Yannakoudakis , Ekaterina Shutova

The Web is naturally heterogeneous with user devices, geographic regions, browsing patterns, and contexts all leading to highly diverse, unique datasets. Federated Learning (FL) is an important paradigm for the Web because it enables…

Machine Learning · Computer Science 2026-02-05 Abdulrahman Alotaibi , Irene Tenison , Miriam Kim , Isaac Lee , Lalana Kagal

In this work, we introduce Adapt & Align, a method for continual learning of neural networks by aligning latent representations in generative models. Neural Networks suffer from abrupt loss in performance when retrained with additional…

Machine Learning · Computer Science 2023-12-22 Kamil Deja , Bartosz Cywiński , Jan Rybarczyk , Tomasz Trzciński

Model merging has emerged as a cost-efficient approximation to multitask learning. Among merging strategies, task arithmetic is notable for its simplicity and effectiveness. In this work, we provide a theoretical motivation for task vectors…

We propose a hybrid approach aimed at improving the sample efficiency in goal-directed reinforcement learning. We do this via a two-step mechanism where firstly, we approximate a model from Model-Free reinforcement learning. Then, we…

Machine Learning · Computer Science 2019-01-09 Shoubhik Debnath , Gaurav Sukhatme , Lantao Liu

Lifelong learning aims to train a model with good performance for new tasks while retaining the capacity of previous tasks. However, some practical scenarios require the system to forget undesirable knowledge due to privacy issues, which is…

Artificial Intelligence · Computer Science 2024-05-30 Lianlei Shan , Wenzhang Zhou , Wei Li , Xingyu Ding

Current training regimes for deep learning usually involve exposure to a single task / dataset at a time. Here we start from the observation that in this context the trained model is not given any knowledge of anything outside its…

Artificial Intelligence · Computer Science 2020-02-11 Giacomo Spigler

Recent advances in Multimodal Large Language Models (MLLMs) have enhanced their versatility as they integrate a growing number of modalities. Considering the heavy cost of training MLLMs, it is efficient to reuse the existing ones and…

Machine Learning · Computer Science 2025-10-23 Dingkun Zhang , Shuhan Qi , Xinyu Xiao , Kehai Chen , Xuan Wang