We report a structural mismatch between a data point’s {learnability}—how quickly it improves the loss—and its {forgettability}—how much it anchors the final parameters—an aspect ignored by prior machine unlearning frameworks such as SISA, Fisher-Forget, and influence-based fine-tuning. To make this gap measurable we introduce Unlearning Gradient Sensitivity (UGS), an influence score computable with a single Hutch++ sketch, and derive the Learnability–Forgettability Divergence (LFD), the Jensen–Shannon distance between the model’s learning and forgetting distributions. We prove that UGS dispersion decays exponentially only under explicit regularisation and that LFD converges to zero when its weight grows sub-linearly relative to the UGS term. Building on these findings, we introduce Dual-Aware Training (DAT)—a lightweight regularization method that reduces variability in how easily data points can be forgotten and aligns learning and forgetting behaviors during training. On CIFAR-10, MNIST, and IMDB, DAT maintains the original model accuracy while cutting forgettability divergence in half and significantly lowering the cost of certified unlearning, showing that it’s effective to make models forgettable from the start.

Federated Transfer Learning (FTL) is a popular approach to solve the problem of heterogeneous feature space and label distribution. Among the mainstream strategies for FTL, parameter decoupling, which balance the impact of a single global model and multiple personalized models under data heterogeneity, has attracted the attention of many researchers. However, few attacks have been proposed to evaluate the privacy risk of FTL. We find that the fine-tuned structures and the gradient update mechanisms of parameter decoupling would be more likely to leak personalized information for the server to infer private labels. Based on our findings, we propose the label inference attack that combines meta classifier with contrastive learning in FTL. Our experiments show that the proposed attack has ability to extract local personalized information from the differences before and after fine-tuning to improve the accuracy of the attack in the absence of a downstream model. Our research can reveal potential privacy risks in FTL and motivate more research on private and secure FTL.

Recent advancements in 3D generation are predominantly propelled by improvements in 3D-aware image diffusion models. These models are pretrained on Internet-scale image data and fine-tuned on massive 3D data, offering the capability of producing highly consistent multi-view images. However, due to the scarcity of synchronized multi-view video data, it remains challenging to adapt this paradigm to 4D generation directly. Despite that, the available video and 3D data are adequate for training video and multi-view diffusion models separately that can provide satisfactory dynamic and geometric priors respectively. To take advantage of both, this paper presents Diffusion$^2$, a novel framework for dynamic 3D content creation that reconciles the knowledge about geometric consistency and temporal smoothness from these models to directly sample dense multi-view multi-frame images which can be employed to optimize continuous 4D representation. Specifically, we design a simple yet effective denoising strategy via score composition of pretrained video and multi-view diffusion models based on the probability structure of the target image array. To alleviate the potential conflicts between two heterogeneous scores, we further introduce variance-reducing sampling via interpolated steps, facilitating smooth and stable generation. Owing to the high parallelism of the proposed image generation process and the efficiency of the modern 4D reconstruction pipeline, our framework can generate 4D content within few minutes. Notably, our method circumvents the reliance on expensive and hard-to-scale 4D data, thereby having the potential to benefit from the scaling of the foundation video and multi-view diffusion models. Extensive experiments demonstrate the efficacy of our proposed framework in generating highly seamless and consistent 4D assets under various types of conditions.

Multi-modal time series analysis has recently emerged as a prominent research area, driven by the increasing availability of diverse data modalities, such as text, images, and structured tabular data from real-world sources. However, effective analysis of multi-modal time series is hindered by data heterogeneity, modality gap, misalignment, and inherent noise. Recent advancements in multi-modal time series methods have exploited the multi-modal context via cross-modal interactions based on deep learning methods, significantly enhancing various downstream tasks. In this tutorial and survey, we present a systematic and up-to-date overview of multi-modal time series datasets and methods. We first state the existing challenges of multi-modal time series analysis and our motivations, with a brief introduction of preliminaries. Then, we summarize the general pipeline and categorize existing methods through a unified cross-modal interaction framework encompassing fusion, alignment, and transference at different levels (i.e., input, intermediate, output), where key concepts and ideas are highlighted. We also discuss the real-world applications of multi-modal analysis for both standard and spatial time series, tailored to general and specific domains. Finally, we discuss future research directions to help practitioners explore and exploit multi-modal time series. The up-to-date resources are provided in the GitHub repository. https://github.com/UConn-DSIS/Multi-modal-Time-Series-Analysis.

Large Language Models (LLMs) and Foundation Models (FMs) have recently become prevalent for time series forecasting tasks. While fine-tuning LLMs enables domain adaptation, they often struggle to generalize across diverse and unseen datasets. Moreover, existing Time Series Foundation Models (TSFMs) still face challenges in handling non-stationary dynamics and distribution shifts, largely due to the lack of effective mechanisms for adaptation. To this end, we present TS-RAG, a retrieval-augmented generation framework for time series forecasting that enhances the generalization and interpretability of TSFMs. Specifically, TS-RAG leverages pre-trained time series encoders to retrieve semantically relevant segments from a dedicated knowledge base, enriching the contextual representation of the input query. Furthermore, we propose an Adaptive Retrieval Mixer (ARM) module that dynamically fuses the retrieved patterns with the TSFM's internal representation, improving forecasting accuracy without requiring task-specific fine-tuning. Thorough empirical studies on seven public benchmark datasets demonstrate that TS-RAG achieves state-of-the-art zero-shot forecasting performance, outperforming the existing TSFMs by up to 6.84\% across diverse domains while also providing desirable interpretability. Our code and data are available at: https://github.com/UConn-DSIS/TS-RAG.

Federated learning (FL) has gained significant traction as a paradigm for decentralized learning, enabling multiple clients to collaboratively train models without sharing their local data. However, applying FL to graph-structured data introduces unique challenges, such as handling non-IID data and preserving the structural dependencies between nodes. Additionally, existing approaches to federated learning with Graph Neural Networks (GNNs) often struggle to capture complex relationships within graph data and are vulnerable to privacy breaches, including membership inference and model inversion attacks. In this paper, we propose Federated Capsule Graph Neural Networks (FCGNN), a novel architecture that integrates the dynamic routing capabilities of capsule networks with the structure-preserving power of GNNs in a federated setting. FCGNN is designed to effectively model hierarchical and part-whole relationships within graph data, enabling it to outperform traditional federated GNN approaches. We enhance the privacy of FCGNN by incorporating differential privacy and secure aggregation techniques, ensuring that individual client updates remain confidential while maintaining strong model performance. We evaluate FCGNN on several benchmark graph datasets, including Cora, Citeseer, PubMed, and PROTEINS, and demonstrate that it consistently achieves higher accuracy and F1-scores compared to existing FL methods. Our experiments show that FCGNN converges faster and incurs lower communication costs, making it highly efficient for real-world FL applications. Furthermore, FCGNN is robust across different numbers of participating clients, maintaining high performance even in non-IID scenarios. These results highlight the potential of FCGNN as a scalable and privacy-preserving solution for decentralized learning on graph-structured data.

In the era of the Internet of Things (IoT), managing the deluge of data generated by distributed devices presents unique challenges, particularly concerning privacy and the efficient use of computational resources. Vertical Federated Learning (VFL) offers a promising avenue for collaborative machine learning without centralizing data, thereby addressing privacy concerns inherent in traditional approaches. However, as data privacy laws and personal data deletion requests become more prevalent, the necessity for effective machine unlearning strategies within VFL frameworks grows increasingly important. To this end, this paper introduces a novel approach to feature-based machine unlearning tailored specifically for VFL systems in IoT networks. Our methodology enables the selective removal of data influence from trained models without the need for full retraining, thus preserving model utility while ensuring compliance with privacy requirements. By integrating a combination of feature relevance measuring techniques and efficient communication protocols, our solution minimizes the data footprint on network nodes, reduces bandwidth consumption, and maintains the integrity and performance of the learning models. To the best of our knowledge, our proposed framework represents the first practical approach to enable machine unlearning within vertical federated learning environments. We demonstrate the effectiveness of our approach through rigorous evaluation using several IoT datasets, highlighting significant improvements in unlearning efficiency and model robustness compared to existing techniques. Our work not only furthers the development of sustainable and compliant machine learning models in IoT but also sets a foundational framework for future research in secure and efficient data management within federated environments.

High-resolution 3D object generation remains a challenging task primarily due to the limited availability of comprehensive annotated training data. Recent advancements have aimed to overcome this constraint by harnessing image generative models, pretrained on extensive curated web datasets, using knowledge transfer techniques like Score Distillation Sampling (SDS). Efficiently addressing the requirements of high-resolution rendering often necessitates the adoption of latent representation-based models, such as the Latent Diffusion Model (LDM). In this framework, a significant challenge arises: To compute gradients for individual image pixels, it is necessary to backpropagate gradients from the designated latent space through the frozen components of the image model, such as the VAE encoder used within LDM. However, this gradient propagation pathway has never been optimized, remaining uncontrolled during training. We find that the unregulated gradients adversely affect the 3D model's capacity in acquiring texture-related information from the image generative model, leading to poor quality appearance synthesis. To address this overarching challenge, we propose an innovative operation termed Pixel-wise Gradient Clipping (PGC) designed for seamless integration into existing 3D generative models, thereby enhancing their synthesis quality. Specifically, we control the magnitude of stochastic gradients by clipping the pixel-wise gradients efficiently, while preserving crucial texture-related gradient directions. Despite this simplicity and minimal extra cost, extensive experiments demonstrate the efficacy of our PGC in enhancing the performance of existing 3D generative models for high-resolution object rendering.

Reconstructing dynamic 3D scenes from 2D images and generating diverse views over time is challenging due to scene complexity and temporal dynamics. Despite advancements in neural implicit models, limitations persist: (i) Inadequate Scene Structure: Existing methods struggle to reveal the spatial and temporal structure of dynamic scenes from directly learning the complex 6D plenoptic function. (ii) Scaling Deformation Modeling: Explicitly modeling scene element deformation becomes impractical for complex dynamics. To address these issues, we consider the spacetime as an entirety and propose to approximate the underlying spatio-temporal 4D volume of a dynamic scene by optimizing a collection of 4D primitives, with explicit geometry and appearance modeling. Learning to optimize the 4D primitives enables us to synthesize novel views at any desired time with our tailored rendering routine. Our model is conceptually simple, consisting of a 4D Gaussian parameterized by anisotropic ellipses that can rotate arbitrarily in space and time, as well as view-dependent and time-evolved appearance represented by the coefficient of 4D spherindrical harmonics. This approach offers simplicity, flexibility for variable-length video and end-to-end training, and efficient real-time rendering, making it suitable for capturing complex dynamic scene motions. Experiments across various benchmarks, including monocular and multi-view scenarios, demonstrate our 4DGS model's superior visual quality and efficiency.

Recently, there has been a growing interest in leveraging pre-trained large language models (LLMs) for various time series applications. However, the semantic space of LLMs, established through the pre-training, is still underexplored and may help yield more distinctive and informative representations to facilitate time series forecasting. To this end, we propose Semantic Space Informed Prompt learning with LLM ($S^2$IP-LLM) to align the pre-trained semantic space with time series embedding space and perform time series forecasting based on learned prompts from the joint space. We first design a tokenization module tailored for cross-modality alignment, which explicitly concatenates patches of decomposed time series components to create embeddings that effectively encode the temporal dynamics. Next, we leverage the pre-trained word token embeddings to derive semantic anchors and align selected anchors with time series embeddings by maximizing the cosine similarity in the joint space. This way, $S^2$IP-LLM can retrieve relevant semantic anchors as prompts to provide strong indicators (context) for time series that exhibit different temporal dynamics. With thorough empirical studies on multiple benchmark datasets, we demonstrate that the proposed $S^2$IP-LLM can achieve superior forecasting performance over state-of-the-art baselines. Furthermore, our ablation studies and visualizations verify the necessity of prompt learning informed by semantic space.

3D representation is essential to the significant advance of 3D generation with 2D diffusion priors. As a flexible representation, NeRF has been first adopted for 3D representation. With density-based volumetric rendering, it however suffers both intensive computational overhead and inaccurate mesh extraction. Using a signed distance field and Marching Tetrahedra, DMTet allows for precise mesh extraction and real-time rendering but is limited in handling large topological changes in meshes, leading to optimization challenges. Alternatively, 3D Gaussian Splatting (3DGS) is favored in both training and rendering efficiency while falling short in mesh extraction. In this work, we introduce a novel 3D representation, Tetrahedron Splatting (TeT-Splatting), that supports easy convergence during optimization, precise mesh extraction, and real-time rendering simultaneously. This is achieved by integrating surface-based volumetric rendering within a structured tetrahedral grid while preserving the desired ability of precise mesh extraction, and a tile-based differentiable tetrahedron rasterizer. Furthermore, we incorporate eikonal and normal consistency regularization terms for the signed distance field to improve generation quality and stability. Critically, our representation can be trained without mesh extraction, making the optimization process easier to converge. Our TeT-Splatting can be readily integrated in existing 3D generation pipelines, along with polygonal mesh for texture optimization. Extensive experiments show that our TeT-Splatting strikes a superior tradeoff among convergence speed, render efficiency, and mesh quality as compared to previous alternatives under varying 3D generation settings.

Federated learning is a distributed machine learning approach that enables multiple participants to collaboratively train a model without sharing their data, thus preserving privacy. However, the decentralized nature of federated learning also makes it susceptible to backdoor attacks, where malicious participants can embed hidden vulnerabilities within the model. Addressing these threats efficiently and effectively is crucial, especially given the impracticality of iterative and resource-intensive detection methods in federated learning environments. This article presents a novel framework for one-shot backdoor removal in federated learning. Our approach integrates advanced anomaly detection techniques with a unique model update aggregation strategy, allowing for the identification and neutralization of backdoor influences in a single update cycle without the need for extensive data access or communication between participants. Extensive experiments across various federated architectures and data distributions demonstrate that our method effectively mitigates backdoor threats while maintaining model performance and scalability. This work not only enhances the security of federated models but also contributes to the broader applicability of federated learning in sensitive and critical domains.

Jointly learning from multiple datasets can help building versatile intelligent systems yet may give rise to serious concerns of data privacy and model selection. Specifically, on the one hand, these datasets can be distributed at various local clients, who may not be willing or do not ought to share data with each other. On the other hand, it is unrealistic to choose a model architecture that can well suit the disparate patterns and distributions carried by the various datasets in a priori. Whereas many works in federated learning [1] and neural architecture search [2] have been proposed to address one of the two concerns, very few have attempted the both. To close the gap, in this paper we deliver a framework, termedMulti-Granular Federated Neural Architecture Search(MGFNAS), to enable the automation of model architecture search in a federated and thus privacy-preserved setting. We argue that our MGFNAS framework is general in the sense that it does not impose any restriction on the search space or strategy, such that most existing neural architecture search techniques can be readily implemented in. The main idea of our framework is to search the optimal neural network architecture in two levels of granularity, enabling the neural-operator-basedmicro-levelsearch and the cell-basedmacro-levelsearch. The main challenge of implementing our framework lies in the fact that, due to the decentralized nature, the local architectures searched by multiple clients can differ drastically in order to fit their own datasets, while a general method to form the global model by aggregating the local architectures in both micro and macro levels is missing. To solve the issue, we propose a novel aggregation function, named Network Architecture Probabilistic Aggregation (NAPA). The key idea of our NAPA function is to treat the network architectures as graphs, of which the sub-graph structures being frequently appeared across multiple clients are modeled by probabilistic distributions. At each round, a global model is formed by sampling from those distributions in an exploration-exploitation fashion. Extensive experiments are carried out, and the results substantiate the viability and effectiveness of our proposed framework.