Few-shot image generation seeks to generate more data of a given domain, with only few available training examples. As it is unreasonable to expect to fully infer the distribution from just a few observations (e.g., emojis), we seek to leverage a large, related source domain as pretraining (e.g., human faces). Thus, we wish to preserve the diversity of the source domain, while adapting to the appearance of the target. We adapt a pretrained model, without introducing any additional parameters, to the few examples of the target domain. Crucially, we regularize the changes of the weights during this adaptation, in order to best preserve the information of the source dataset, while fitting the target. We demonstrate the effectiveness of our algorithm by generating high-quality results of different target domains, including those with extremely few examples (e.g., 10). We also analyze the performance of our method with respect to some important factors, such as the number of examples and the similarity between the source and target domain.

Knowledge graphs (KGs) require continual updates as new information emerges, but neural embedding models suffer from catastrophic forgetting when learning new tasks sequentially. We evaluate Elastic Weight Consolidation (EWC), a regularization-based continual learning method, on KG link prediction using TransE embeddings on FB15k-237. Across multiple experiments with five random seeds, we find that EWC reduces catastrophic forgetting from 12.62% to 6.85%, a 45.7% reduction compared to naive sequential training. We observe that the task partitioning strategy affects the magnitude of forgetting: relation-based partitioning (grouping triples by relation type) exhibits 9.8 percentage points higher forgetting than randomly partitioned tasks (12.62% vs 2.81%), suggesting that task construction influences evaluation outcomes. While focused on a single embedding model and dataset, our results demonstrate that EWC effectively mitigates catastrophic forgetting in KG continual learning and highlight the importance of evaluation protocol design.

Medical imaging foundation models must adapt over time, yet full retraining is often blocked by privacy constraints and cost. We present a continual learning framework that avoids storing patient exemplars by pairing class conditional diffusion replay with Elastic Weight Consolidation. Using a compact Vision Transformer backbone, we evaluate across eight MedMNIST v2 tasks and CheXpert. On CheXpert our approach attains 0.851 AUROC, reduces forgetting by more than 30\% relative to DER\texttt{++}, and approaches joint training at 0.869 AUROC, while remaining efficient and privacy preserving. Analyses connect forgetting to two measurable factors: fidelity of replay and Fisher weighted parameter drift, highlighting the complementary roles of replay diffusion and synaptic stability. The results indicate a practical route for scalable, privacy aware continual adaptation of clinical imaging models.

Graph continual learning (GCL) aims to learn from a continuous sequence of graph-based tasks. Regularization methods are vital for preventing catastrophic forgetting in GCL, particularly in the challenging replay-free, class-incremental setting, where each task consists of a set of unique classes. In this work, we first establish a general regularization framework for GCL based on the curved parameter space induced by the Fisher information matrix (FIM). We show that the dominant Elastic Weight Consolidation (EWC) and its variants are a special case within this framework, using a diagonal approximation of the empirical FIM based on parameters from previous tasks. To overcome their limitations, we propose a new unbiased online curvature approximation of the full FIM based on the model's current learning state. Our method directly estimates the regularization term in an online manner without explicitly evaluating and storing the FIM itself. This enables the model to better capture the loss landscape during learning new tasks while retaining the knowledge learned from previous tasks. Extensive experiments on three graph datasets demonstrate that our method significantly outperforms existing regularization-based methods, achieving a superior trade-off between stability (retaining old knowledge) and plasticity (acquiring new knowledge).

Crucially, we regularize the changes of the weights during this adaptation, in order to best preserve the "information" of the source dataset, while fitting the target. We demonstrate the effectiveness of our algorithm by generating high-quality results of different target domains, including those with extremely few examples (e.g.,

Continual learning systems face the dual challenge of preventing catastrophic forgetting while maintaining energy efficiency, particularly in resource-constrained environments. This paper introduces Pathway-based Progressive Inference (PaPI), a novel theoretical framework that addresses these challenges through a mathematically rigorous approach to pathway selection and adaptation. We formulate continual learning as an energy-constrained optimization problem and provide formal convergence guarantees for our pathway routing mechanisms. Our theoretical analysis demonstrates that PaPI achieves an $\mathcal{O}(K)$ improvement in the stability-plasticity trade-off compared to monolithic architectures, where $K$ is the number of pathways. We derive tight bounds on forgetting rates using Fisher Information Matrix analysis and prove that PaPI's energy consumption scales with the number of active parameters rather than the total model size. Comparative theoretical analysis shows that PaPI provides stronger guarantees against catastrophic forgetting than Elastic Weight Consolidation (EWC) while maintaining better energy efficiency than both EWC and Gradient Episodic Memory (GEM). Our experimental validation confirms these theoretical advantages across multiple benchmarks, demonstrating PaPI's effectiveness for continual learning in energy-constrained settings. Our codes are available at https://github.com/zser092/PAPI_FILES.

Weaknesses: The quality of the paper is already great, but there are a few comments. In equation 3 (page 4), it is not clear whether you compute F of the generated source or target data. Also, I don't quite understand why the FI is computed for the difference between the pretrained and finetuned parameters, and not just for the pretrained parameters. Finally, I assume i in this equation is the layer index, but this should be clearly stated. Update: In the rebuttal, the authors kindly explained that the F is computed for each individual parameter in the network rather than for the entire layer.

This paper proposes an interesting framework for generating images from few-shot data. While some reviewers were concerned about the quality of generated images, the ideas in this paper is interesting enough to justify publicaition.

Few-shot image generation seeks to generate more data of a given domain, with only few available training examples. As it is unreasonable to expect to fully infer the distribution from just a few observations (e.g., emojis), we seek to leverage a large, related source domain as pretraining (e.g., human faces). Thus, we wish to preserve the diversity of the source domain, while adapting to the appearance of the target. We adapt a pretrained model, without introducing any additional parameters, to the few examples of the target domain. Crucially, we regularize the changes of the weights during this adaptation, in order to best preserve the information of the source dataset, while fitting the target. We demonstrate the effectiveness of our algorithm by generating high-quality results of different target domains, including those with extremely few examples (e.g., 10).

Sequential learning involves learning tasks in a sequence, and proves challenging for most neural networks. Biological neural networks regularly conquer the sequential learning challenge and are even capable of transferring knowledge both forward and backwards between tasks. Artificial neural networks often totally fail to transfer performance between tasks, and regularly suffer from degraded performance or catastrophic forgetting on previous tasks. Models of associative memory have been used to investigate the discrepancy between biological and artificial neural networks due to their biological ties and inspirations, of which the Hopfield network is perhaps the most studied model. The Dense Associative Memory, or modern Hopfield network, generalizes the Hopfield network, allowing for greater capacities and prototype learning behaviors, while still retaining the associative memory structure. We investigate the performance of the Dense Associative Memory in sequential learning problems, and benchmark various sequential learning techniques in the network. We give a substantial review of the sequential learning space with particular respect to the Hopfield network and associative memories, as well as describe the techniques we implement in detail. We also draw parallels between the classical and Dense Associative Memory in the context of sequential learning, and discuss the departures from biological inspiration that may influence the utility of the Dense Associative Memory as a tool for studying biological neural networks. We present our findings, and show that existing sequential learning methods can be applied to the Dense Associative Memory to improve sequential learning performance.