English
Related papers

Related papers: Grow, Don't Overwrite: Fine-tuning Without Forgett…

200 papers

In real-world applications, learning-enabled systems often undergo iterative model development to address challenging or emerging tasks, which involve collecting new data, training a new model and validating the model. This continual model…

Machine Learning · Computer Science 2025-04-22 Gang Li , Wendi Yu , Yao Yao , Wei Tong , Yingbin Liang , Qihang Lin , Tianbao Yang

Lifelong learning aims to develop machine learning systems that can learn new tasks while preserving the performance on previous learned tasks. In this paper we present a method to overcome catastrophic forgetting on convolutional neural…

Machine Learning · Computer Science 2018-05-10 Abel S. Zacarias , Luís A. Alexandre

Using neural networks in practical settings would benefit from the ability of the networks to learn new tasks throughout their lifetimes without forgetting the previous tasks. This ability is limited in the current deep neural networks by a…

Machine Learning · Computer Science 2018-06-20 Risto Vuorio , Dong-Yeon Cho , Daejoong Kim , Jiwon Kim

Deep neural networks have revolutionized numerous research fields and applications. Despite their widespread success, a fundamental limitation known as catastrophic forgetting remains, where models fail to retain their ability to perform…

Machine Learning · Computer Science 2025-02-21 Karolina Bogacka , Maximilian Höfler , Maria Ganzha , Wojciech Samek , Katarzyna Wasielewska-Michniewska

A well-trained Convolutional Neural Network can easily be pruned without significant loss of performance. This is because of unnecessary overlap in the features captured by the network's filters. Innovations in network architecture such as…

Computer Vision and Pattern Recognition · Computer Science 2019-02-26 Aaditya Prakash , James Storer , Dinei Florencio , Cha Zhang

Catastrophic forgetting is a pervasive issue for pre-trained language models (PLMs) during continual learning, where models lose previously acquired knowledge when sequentially trained on a series of tasks. The model's ability to retain old…

Computation and Language · Computer Science 2025-02-18 Biqing Zeng , Zehan Li , Aladdin Ayesh

This paper considers continual learning of large-scale pretrained neural machine translation model without accessing the previous training data or introducing model separation. We argue that the widely used regularization-based methods,…

Computation and Language · Computer Science 2022-11-07 Shuhao Gu , Bojie Hu , Yang Feng

When incrementally trained on new classes, deep neural networks are subject to catastrophic forgetting which leads to an extreme deterioration of their performance on the old classes while learning the new ones. Using a small memory…

Machine Learning · Computer Science 2022-11-15 Quentin Jodelet , Xin Liu , Tsuyoshi Murata

Large language models (LLMs) have achieved significant progress from pre-training on and memorizing a wide range of textual data, however, this process might suffer from privacy issues and violations of data protection regulations. As a…

Computation and Language · Computer Science 2023-11-01 Jiaao Chen , Diyi Yang

Matching animal-like flexibility in recognition and the ability to quickly incorporate new information remains difficult. Limits are yet to be adequately addressed in neural models and recognition algorithms. This work proposes a…

Computer Vision and Pattern Recognition · Computer Science 2012-06-26 Tsvi Achler

Large pre-trained language models (PLMs) have demonstrated strong performance on natural language understanding (NLU) tasks through fine-tuning. However, fine-tuned models still suffer from overconfident predictions, especially in…

Computation and Language · Computer Science 2023-05-31 Guande He , Jianfei Chen , Jun Zhu

This paper explores the possibility of extending the capability of pre-trained neural image compressors (e.g., adapting to new data or target bitrates) without breaking backward compatibility, the ability to decode bitstreams encoded by the…

Image and Video Processing · Electrical Eng. & Systems 2024-03-01 Zhihao Duan , Ming Lu , Justin Yang , Jiangpeng He , Zhan Ma , Fengqing Zhu

Large language models (LLMs) excel in various tasks but are primarily trained on text data, limiting their application scope. Expanding LLM capabilities to include vision-language understanding is vital, yet training them on multimodal data…

Computer Vision and Pattern Recognition · Computer Science 2024-08-08 Shanshan Zhong , Shanghua Gao , Zhongzhan Huang , Wushao Wen , Marinka Zitnik , Pan Zhou

Pretrained language models (PLMs) are today the primary model for natural language processing. Despite their impressive downstream performance, it can be difficult to apply PLMs to new languages, a barrier to making their capabilities…

Computation and Language · Computer Science 2024-01-15 Yihong Chen , Kelly Marchisio , Roberta Raileanu , David Ifeoluwa Adelani , Pontus Stenetorp , Sebastian Riedel , Mikel Artetxe

Post-training for large language models (LLMs) is constrained by the high cost of acquiring new knowledge or correcting errors and by the unintended side effects that frequently arise from retraining. To address these issues, we introduce…

Computation and Language · Computer Science 2026-02-11 Yisu Wang , Ming Wang , Haoyuan Song , Wenjie Huang , Chaozheng Wang , Yi Xie , Xuming Ran

Multitask learning assumes that models capable of learning from multiple tasks can achieve better quality and efficiency via knowledge transfer, a key feature of human learning. Though, state of the art ML models rely on high customization…

Machine Learning · Computer Science 2022-11-17 Andrea Gesmundo , Jeff Dean

Federated fine-tuning (FedFT) of large language models (LLMs) has emerged as a promising solution for adapting models to distributed data environments while ensuring data privacy. Existing FedFT methods predominantly utilize…

Machine Learning · Computer Science 2025-06-09 Yujia Huo , Jianchun Liu , Hongli Xu , Zhenguo Ma , Shilong Wang , Liusheng Huang

Large language models store all learned knowledge in a single, fixed weight vector. Teaching a model new capabilities requires modifying those same weights, inevitably degrading previously acquired knowledge. This fundamental limitation,…

Machine Learning · Computer Science 2026-02-24 Sarim Chaudhry

Currently used semantic parsing systems deployed in voice assistants can require weeks to train. Datasets for these models often receive small and frequent updates, data patches. Each patch requires training a new model. To reduce training…

Computation and Language · Computer Science 2021-03-23 Vladislav Lialin , Rahul Goel , Andrey Simanovsky , Anna Rumshisky , Rushin Shah

Rehearsal is one of the key techniques for mitigating catastrophic forgetting and has been widely adopted in continual learning algorithms due to its simplicity and practicality. However, the theoretical understanding of how rehearsal scale…

Machine Learning · Computer Science 2026-02-25 JinLi He , Liang Bai , Xian Yang