多篇研究论文探讨了持续学习的先进技术，旨在防止AI模型的灾难性遗忘。一种方法“经验混合”（Experience Blending, EB）使用生成的“支持边界数据”来丰富决策边界并提高图像分类等任务的准确性。另一种方法TeLAPA通过组织多样化的策略邻域而非依赖单一演进策略来解决持续强化学习问题，从而增强适应性和重用性。其他研究则探讨了任务粒度对遗忘的影响，提出了用于监控大型预训练模型在线学习的LargeMonitor等框架，并引入了用于调节适应性的Drift-Plus-Penalty等理论基础。
AI
arXiv:2606.13258v1 Announce Type: new Abstract: Gait-based Parkinson's disease assessment increasingly relies on heterogeneous sensors, but clinical systems rarely collect all modalities simultaneously. New sensors may arrive through device upgrades, protocol changes, or multi-ce…
Catastrophic forgetting is often viewed as the destruction of previously learned knowledge during sequential learning. Building on the Accessibility Collapse framework, we investigate the geometric structure of recoverability in continual learning. Using Split CIFAR-100 and a seq…
Gait-based Parkinson's disease assessment increasingly relies on heterogeneous sensors, but clinical systems rarely collect all modalities simultaneously. New sensors may arrive through device upgrades, protocol changes, or multi-center deployment, while historical patient data a…
arXiv:2507.23534v3 Announce Type: replace Abstract: Continual learning (CL) seeks to mitigate catastrophic forgetting when models are trained with sequential tasks. A common approach, experience replay (ER), stores past exemplars but only sparsely approximates the data distributi…
arXiv cs.AI
TIER_1English(EN)·Lute Lillo, Nick Cheney·
arXiv:2604.15414v2 Announce Type: replace-cross Abstract: Continual reinforcement learning must balance retention with adaptation, yet many methods still rely on \emph{single-model preservation}, committing to one evolving policy as the main reusable solution across tasks. Even w…
arXiv:2606.08013v1 Announce Type: new Abstract: Catastrophic forgetting, the abrupt loss of previously acquired knowledge upon learning new information, remains the central challenge in Continual Learning. This project investigates whether the order in which a model learns inform…
arXiv:2606.09430v1 Announce Type: cross Abstract: Online task-free continual learning (TFCL) requires intelligent agents to sequentially accumulate knowledge from an unbounded, non-stationary data stream under strict single-pass constraints and without any explicit task identifie…
arXiv:2606.08452v1 Announce Type: new Abstract: In many real-world settings, data streams are nonstationary and arrive sequentially, requiring learning systems to adapt continuously without retraining from scratch. Continual learning (CL) addresses this challenge by incorporating…
arXiv cs.AI
TIER_1English(EN)·Andries Rosseau, Robert M\"uller, Ann Now\'e·
arXiv:2606.09762v1 Announce Type: cross Abstract: Continual training of deep neural networks under non-stationarity often leads to a progressive loss of plasticity, eventually limiting further learning. We relate plasticity to the empirical Neural Tangent Kernel, and identify dyn…
Continual training of deep neural networks under non-stationarity often leads to a progressive loss of plasticity, eventually limiting further learning. We relate plasticity to the empirical Neural Tangent Kernel, and identify dynamical isometry (the condition that layer-wise Jac…
Continual training of deep neural networks under non-stationarity often leads to a progressive loss of plasticity, eventually limiting further learning. We relate plasticity to the empirical Neural Tangent Kernel, and identify dynamical isometry (the condition that layer-wise Jac…
Online task-free continual learning (TFCL) requires intelligent agents to sequentially accumulate knowledge from an unbounded, non-stationary data stream under strict single-pass constraints and without any explicit task identifiers. Existing online TFCL paradigms primarily rely …
arXiv:2606.07500v1 Announce Type: cross Abstract: Continual learning in Large Language Models (LLMs) is hindered by the plasticity-stability dilemma, where acquiring new capabilities often leads to catastrophic forgetting of previous knowledge. Existing methods typically treat pa…
arXiv:2505.12239v2 Announce Type: replace-cross Abstract: In Continual Learning (CL), using a Pre-Trained Model (PTM) as the feature extractor has become a popular practice. Accompanied by analytic classifiers, the PTM-based methods have achieved state-of-the-art performance in C…
In many real-world settings, data streams are nonstationary and arrive sequentially, requiring learning systems to adapt continuously without retraining from scratch. Continual learning (CL) addresses this challenge by incorporating new tasks while mitigating catastrophic forgett…
Continual learning in Large Language Models (LLMs) is hindered by the plasticity-stability dilemma, where acquiring new capabilities often leads to catastrophic forgetting of previous knowledge. Existing methods typically treat parameters uniformly, failing to distinguish between…
arXiv:2606.06032v1 Announce Type: new Abstract: Catastrophic forgetting is commonly interpreted as the irreversible erasure of previously acquired knowledge during sequential learning. In this work, we investigate an alternative perspective: that forgetting may arise not from com…
arXiv:2606.05695v1 Announce Type: new Abstract: Exemplar-free class-incremental learning (EFCIL) aims to acquire new classes over time without storing raw data. Historically, prototype rehearsal, which samples around stored class prototypes and mixes them with current-task data, …
arXiv:2606.05559v1 Announce Type: new Abstract: Deployed large language model agents must adapt to distribution shift in dynamic environments. Ideally, adaptation can be performed from accumulated agent experiences and retain prior capabilities while transferring to future tasks.…
arXiv cs.CL
TIER_1English(EN)·Parth Asawa, Christopher M. Glaze, Gabriel Orlanski, Ramya Ramakrishnan, Benji Xu, Asim Biswal, Vincent Sunn Chen, Frederic Sala, Matei Zaharia, Joseph E. Gonzalez·
arXiv:2606.05661v1 Announce Type: cross Abstract: Continual learning, the ability of AI systems to improve through sequential experience, has attracted substantial interest, but no high-quality benchmark exists to evaluate it. We introduce Continual Learning Bench (CL-Bench), the…
arXiv cs.LG
TIER_1English(EN)·Marius Dragoi, Ioana Pintilie, Alexandra Dragomir, Antonio Barbalau, Florin Brad·
arXiv:2606.06494v1 Announce Type: new Abstract: Parameter-efficient finetuning methods based on spectral decomposition have enabled progress in Continual Learning. In this paper we introduce TailLoR, which utilizes the singular bases U and V of the pre-trained weights as a fixed …
Parameter-efficient finetuning methods based on spectral decomposition have enabled progress in Continual Learning. In this paper we introduce TailLoR, which utilizes the singular bases U and V of the pre-trained weights as a fixed reference frame to learn a low-rank update appli…
Catastrophic forgetting is commonly interpreted as the irreversible erasure of previously acquired knowledge during sequential learning. In this work, we investigate an alternative perspective: that forgetting may arise not from complete destruction of task representations but fr…
arXiv cs.AI
TIER_1English(EN)·Amogh Inamdar, Matthew So, Vici Milenia, Richard Zemel·
arXiv:2606.03843v1 Announce Type: cross Abstract: Continual learning methods aim to maximize the stability and plasticity of machine learning models that are trained on a sequence of tasks. The standard measure of stability (i.e., forgetting) is the 0-shot performance of a model …
Continual learning methods aim to maximize the stability and plasticity of machine learning models that are trained on a sequence of tasks. The standard measure of stability (i.e., forgetting) is the 0-shot performance of a model on previously learned tasks, and plasticity, the p…
arXiv:2606.01379v1 Announce Type: new Abstract: While prompt-based parameter-efficient continual learning mitigates catastrophic forgetting by isolating task-specific prompts, this isolation also limits later tasks from improving earlier ones, leaving backward knowledge transfer …
arXiv cs.LG
TIER_1English(EN)·Alexander Chebykin, Tanja Alderliesten, Peter A. N. Bosman·
arXiv:2511.09190v2 Announce Type: replace Abstract: Hyperparameter Optimization (HPO) can lift the burden of tuning hyperparameters (HPs) of neural networks. HPO algorithms from the Population Based Training (PBT) family are efficient thanks to dynamically adjusting HPs every few…
arXiv:2606.01863v1 Announce Type: new Abstract: Continual learning struggles to balance retaining past knowledge with absorbing new tasks. Stefan-CL elegantly resolves this stability-plasticity dilemma through the physics of melting. It frames consolidated knowledge as a protecte…
arXiv:2605.19145v2 Announce Type: replace Abstract: In the literature, many continual learning (CL) algorithms have been proposed to address the issue of catastrophic forgetting in ML models (i.e., learning new tasks leads to the loss of performance on previously learned tasks). …
