论文检索

Nelson Ma, Junyu Xuan, Guangquan Zhang, Jie Lu

Skill discovery has emerged as a popular route for unsupervised reinforcement learning (URL), offering agents a diverse, reusable set of behaviours learned before any task-specific reward is experienced. However, existing methodologies tend to favour either categorical codes or unimodal skill priors, which simplifies training at the cost of limiting the variety of behaviours they can represent. We introduce Discovery of Mixture Skills (DiMS), a URL algorithm that learns a latent Gaussian mixture by training a Gaussian Mixture Variational Autoencoder (GMVAE) in tandem with the unsupervised policy. In DiMS, a hierarchical GMVAE simultaneously discovers clusters of skills, while an auxiliary macro-latent dynamically positions mixture components to prevent mode collapse. A joint loss term combining log-likelihood and curiosity rewards enables simultaneous updates of representation and policy while improving exploration. Experiments on the Unsupervised Reinforcement Learning Benchmark (URLB) show that DiMS consistently outperforms a wide range of state-of-the-art baselines. Ablation studies confirm that the mixture prior is critical to these gains, and that DiMS is robust to alternative exploration bonuses. Overall, our results suggest that Gaussian mixture skill priors offer a compelling foundation for future unsupervised RL.

Yuyuan Li, Junjie Fang, Fengyuan Yu, Xichun Sheng, Tianyu Du, Xuyang Teng, Shaowei Jiang, Linbo Jiang, Jianan Lin, Chaochao Chen

Federated Recommender Systems (FedRecs) leverage federated learning to protect user privacy by retaining data locally. However, user embeddings in FedRecs often encode sensitive attribute information, rendering them vulnerable to attribute inference attacks. Attribute unlearning has emerged as a promising approach to mitigate this issue. In this paper, we focus on user-level FedRecs, which is a more practical yet challenging setting compared to group-level FedRecs. Adversarial training emerges as the most feasible approach within this context. We identify two key challenges in implementing adversarial training-based attribute unlearning for user-level FedRecs: i) mitigating training instability caused by user data heterogeneity, and ii) preventing attribute information leakage through gradients. To address these challenges, we propose FedAU2, an attribute unlearning method for user-level FedRecs. For CH1, we propose a adaptive adversarial training strategy, where the training dynamics are adjusted in response to local optimization behavior. For CH2, we propose a dual-stochastic variational autoencoder to perturb the adversarial model, effectively preventing gradient-based information leakage. Extensive experiments on three real-world datasets demonstrate that our proposed FedAU2 achieves superior performance in unlearning effectiveness and recommendation performance compared to existing baselines.

Yue Li, Weifan Wang, Tai Sing Lee

Recent neurophysiological studies have revealed that the early visual cortex can rapidly learn global image context, as evidenced by a sparsification of population responses and a reduction in mean activity when exposed to familiar versus novel image contexts. This phenomenon has been attributed primarily to local recurrent interactions, rather than changes in feedforward or feedback pathways—supported by both empirical findings and circuit-level modeling. Recurrent neural circuits capable of simulating these effects have been shown to reshape the geometry of neural manifolds, enhancing robustness and invariance to irrelevant variations. In this study, we employ a Vision Transformer (ViT)-based autoencoder to investigate, from a functional perspective, how familiarity training can induce sensitivity to global context in the early layers of a deep neural network. We hypothesize that rapid learning operates via fast weights, which encode transient or short-term memory traces, and we explore the use of Low-Rank Adaptation (LoRA) to implement such fast weights within each Transformer layer. Our results show that: (1) The proposed ViT-based autoencoder's self-attention circuit is performing a manifold transform similar to a neural circuit developed for modeling the familiarity effect. (2) Familiarity training induces alignment of latent representation in early layers with the top layer that contains global context information. (3) Familiarity training makes self-attention pay attention to a broader scope details in the remembered image context, rather than just the critical features for object recognition. (4) These effects are significantly amplified by the incorporation of LoRA-based fast weights. Together, these findings suggest that familiarity training can introduce global sensitivity to earlier layers in a hierarchical network, and that a hybrid fast-and-slow weight architecture may provide a viable computational model for studying the functional consequences of rapid global context learning in the brain.

Haokai Hong, Wanyu Lin, Ming Yang, Kay Chen Tan

Can we train a 3D molecule generator using data from dense regions to generate samples in sparse regions? This challenge can be framed as an out-of-distribution (OOD) generation problem. While prior research on OOD generation predominantly targets property shifts, structural shifts, such as differences in molecular scaffolds or functional groups, represent an equally critical source of distributional shifts. This work introduces the Geometric OOD Diffusion Model (GODD), a novel diffusion-based framework that enables training on data-abundant molecular distributions while generalizing to data-scarce distributions under distributional structural shifts. Central to our approach is a designated equivariant asymmetric autoencoder to capture distributional structural priors. The asymmetric design allows the model to generalize to unseen structural variations by capturing distributional priors representing distinct distributions. The encoded structural-grained priors guide generation toward sparse regions without requiring explicit training on such data. Evaluated across standard benchmarks encompassing OOD structural shifts (e.g., scaffolds, rings), GODD achieves an improvement of 12.6% in success rate, defined based on molecular validity, uniqueness, and novelty. Furthermore, the framework demonstrates promising performance and generalization on canonical fragment-based drug design tasks, highlighting its utility in learning-based molecular discovery.

Xinyue Guo, Xiaoran Yang, Lipan Zhang, Jianxuan Yang, Zhao Wang, Jian Luan

Sound effect editing—modifying audio by adding, removing, or replacing elements—remains constrained by existing approaches that rely solely on low-level signal processing or coarse text prompts, often resulting in limited flexibility and suboptimal audio quality. To address this, we propose AV-Edit, a generative sound effect editing framework that enables fine-grained editing of existing audio tracks in videos by jointly leveraging visual, audio, and text semantics. Specifically, the proposed method employs a specially designed contrastive audio-visual masking autoencoder (CAV-MAE-Edit) for multimodal pre-training, learning aligned cross-modal representations. These representations are then used to train an editorial Multimodal Diffusion Transformer (MM-DiT) capable of removing visually irrelevant sounds and generating missing audio elements consistent with video content through a correlation-based feature gating training strategy. Furthermore, we construct a dedicated video-based sound editing dataset as an evaluation benchmark. Experiments demonstrate that the proposed AV-Edit generates high-quality audio with precise modifications based on visual content, achieving state-of-the-art performance in the field of sound effect editing and exhibiting strong competitiveness in the domain of audio generation.

Alan J.X. Guo, Mengyi Wei, Yufan Dai, Yali Wei, Pengchen Zhang

With recent advancements in next-generation data storage, especially in biological molecule-based storage, insertion, deletion, and substitution (IDS) error-correcting codes have garnered increased attention. However, a universal method for designing tailored IDS-correcting codes across varying channel settings remains underexplored. We present an autoencoder-based approach, THEA-code, aimed at efficiently generating IDS-correcting codes for complex IDS channels. In the work, a disturbance-based discretization is proposed to discretize the features of the autoencoder, and a simulated differentiable IDS channel is developed as a differentiable alternative for IDS operations. These innovations facilitate the successful convergence of the autoencoder, producing channel-customized IDS-correcting codes that demonstrate commendable performance across complex IDS channels, particularly in realistic DNA-based storage channels.

Renxiang Guan, Junhong Li, Siwei Wang, Tianrui Liu, Dayu Hu, Miaomiao Li, Xinwang Liu

Multi-view graph clustering (MVGC) for remote sensing data has gained increasing attention due to its ability to integrate complementary information across modalities while capturing spatial dependencies in heterogeneous data. Although current methods based on graph contrastive learning achieve strong performance, they often misidentify intra-cluster samples as negatives, leading to class conflicts and reduced clustering accuracy. Graph masked autoencoders have recently shown promising potential in learning robust representations through masked reconstruction, but their application to remote sensing data remains underexplored. This challenge is especially notable in the multi-view remote sensing setting, where high heterogeneity and complex spatial structures increase the difficulty of effective representation learning. To address these issues, we propose Clustering-Guided graph Mask AutoEncoder (CG-MAE), the first framework to extend graph masked autoencoders to multi-view remote sensing clustering. We introduce a clustering-guided masking strategy that selectively masks nodes near cluster centers and intra-cluster edges, which are crucial for capturing key structural information. By reconstructing these masked components, the model is encouraged to focus on learning features that are highly relevant to clustering. To further improve training stability and efficiency, we design an easy-to-hard node masking strategy that enables the model to gradually learn from increasingly challenging patterns. Additionally, we propose a dual self-adaptive learning mechanism that encourages the model to align more closely with the underlying semantic distributions. Extensive experiments on four widely used multi-view remote sensing datasets demonstrate that CG-MAE consistently outperforms state-of-the-art methods in both clustering accuracy and representation quality.

Xinshun Feng, Mingzhe Liu, Yi Qiao, Tongyu Zhu, Leilei Sun, Shuai Wang

Recent advances in explainable recommendation have explored the integration of language models to analyze natural language rationales for user–item interactions. Despite their potential, existing methods often rely on ID-based representations that obscure semantic meaning and impose structural constraints on language models, thereby limiting their applicability in open-ended scenarios. These challenges are intensified by the complex nature of real-world interactions, where diverse user intents are entangled and collaborative signals rarely align with linguistic semantics. To overcome these limitations, we propose BEAT, a unified and transferable framework that tokenizes user and item behaviors into discrete, interpretable sequences. We construct a behavior vocabulary via a vector-quantized autoencoding process that disentangles macro-level interests and micro-level intentions from graph-based representations. We then introduce multi-level semantic supervision to bridge the gap between behavioral signals and language space. A semantic alignment regularization mechanism is designed to embed behavior tokens directly into the input space of frozen language models. Experiments on three public datasets show that BEAT improves zero-shot recommendation performance while generating coherent and informative explanations. Further analysis demonstrates that our behavior tokens capture fine-grained semantics and offer a plug-and-play interface for integrating complex behavior patterns into large language models.

Yiming Du, Ziyu Wang, Jian Li, Rui Ning, Lusi Li

Incomplete multi-view clustering (IMVC) aims to discover shared cluster structures from multi-view data with partial observations. The core challenges lie in accurately imputing missing views without introducing bias, while maintaining semantic consistency across views and compactness within clusters. To address these challenges, we propose DIMVC-HIA, a novel deep IMVC framework that integrates hierarchical imputation and alignment with four key components: (1) view-specific autoencoders for latent feature extraction, coupled with a view-shared clustering predictor to produce soft cluster assignments; (2) a hierarchical imputation module that first estimates missing cluster assignments based on cross-view contrastive similarity, and then reconstructs missing features using intra-view, intra-cluster statistics; (3) an energy-based semantic alignment module, which promotes intra-cluster compactness by minimizing energy variance around low-energy cluster anchors; and (4) a contrastive assignment alignment module, which enhances cross-view consistency and encourages confident, well-separated cluster predictions. Experiments on benchmarks demonstrate that our framework achieves superior performance under varying levels of missingness.

Ibrahim Delibasoglu, Sanjay Chakraborty, Fredrik Heintz, Mattias Tiger

Understanding how localized changes in one variable affect others in multivariate time series is essential for diagnostics and decision-making in complex systems. Existing models often fail to capture realistic inter-feature dynamics when simulating "what-if" scenarios, leading to inaccurate or uncorrelated reconstructions. We propose CFORVAE, a variational autoencoder framework that explicitly addresses this limitation by combining temporal decomposition with frequency-domain feature correlation modeling. Our architecture uses a dual-path encoding of trend and seasonal components, each projected into attention-pooled latent spaces, and applies Fourier Neural Operators (FNO) to capture cross-feature dependencies in the spectral domain. This decomposition-correlation design enables component-specific latent manipulation and ensures that local modifications propagate realistically across correlated variables. Through extensive experiments, we show that CFORVAE outperforms state-of-the-art baselines in preserving temporal and feature-level dependencies, especially under adjustment-based reconstructions, making it a powerful tool for interpretable "what-if" analysis and diagnostics.

Zhaoliang Chen, William K. Cheung, Hong-Ning Dai, Byron Choi, Jiming Liu

Multi-view clustering has been found useful to leverage diverse data sources for accurate and robust underlying data representations. It typically relies on effectively integrating the latent features from different views through allocating weights while simultaneously mining their specificity and consensus information. However, it remains open how to achieve a more fine-grained sample-level weight allocation for promoting view-specific information fusion and view-shared consensus. To address this problem, we propose a novel multi-expert learning framework named Gated Variational Graph AutoEncoder with Competition and Consensus (GVGAE-C2). In particular, it employs multiple view-specific Variational Graph AutoEncoders (VGAEs) as experts to capture the latent features from their own views. Furthermore, we design a fine-grained structure-aware gating network, which dynamically computes sample-level weights based on the proposed structure-aware quality evaluation on each expert, thus facilitating competition among experts. Meanwhile, each expert is trained not only to study its assigned view's specificity features, but also explicitly encouraged to learn consensus-aware features across views. Extensive multi-view clustering experiments on benchmark datasets reveal that GVGAE-C2 significantly outperforms state-of-the-art methods.

Xin Chen, Qi Zhao, Wei Zeng, Zongben Xu

Images are generally represented by pixel intensities or color values, which are usually used as direct inputs for learning. This study innovatively proposes a geometric image representation method and refreshes the general learning model (e.g., autoencoder) in the diffeomorphic space. Based on the theory of geometric optimal transport and quasiconformal mapping, we equivalently transform the intensity representation into a shape representation. The image space becomes a diffeomorphic space, where any image can be uniquely represented as a Beltrami coefficient function defined on a uniform grid reference, and vice versa. This innovative geometric image representation (G-IR) captures the fine-grained structure inherent in the entire image, which is different from the traditional feature extraction that focuses on the internal geometric objects of the image (such as boundaries and axes). The diffeomorphic property preserves structure in the generation process, which is very necessary in the field of real physics. It can be assembled into existing pipelines as a plug-in, providing structure-preserving properties for the entire framework. Experiments on image restoration and interpolation validated the high efficiency, efficacy and applicability of the G-IR method, demonstrating its superior performance compared to common pixel-level image appearance representations.

Junyang Chen, Huan Wang, Yirui Wu, Qiuzhen Lin, Yunfeng Diao, Junkai Ji

Micro-video label prediction plays a pivotal role on contemporary video-sharing platforms, such as Kwai and Tiktok. The emergence of video content lacking labels presents a formidable challenge for conventional user interest prediction methods. This paper addresses the challenge of micro-video label prediction, particularly for unseen videos, by proposing a zero-shot method called Class Semantic Relation Learning (CSRL). Unlike traditional user interest prediction models, CSRL leverages the pre-trained Large Language Model (LLM) to enhance prediction accuracy for unlabeled videos. The novelty of CSRL lies in its integration of three key components: a raw feature autoencoder, LLM-enhanced features, and a decomposed graph network. The decomposed graph network is specifically designed to disentangle the relationships between labeled and unlabeled videos, offering a significant improvement over previous methods. By fusing hidden topics with LLM-enhanced text, CSRL effectively handles sparse video features. Experiments on large-scale datasets from the Kwai platform show that CSRL achieves state-of-the-art results, with up to 44.64% improvement in Hit Ratio (HR), highlighting its superiority over existing zero-shot recommendation models in predicting user interests within the user-video network.

Yingji Zhang, Marco Valentino, Danilo Carvalho, André Freitas

Incorporating explicit reasoning rules within the latent space of language models (LMs) offers a promising pathway to enhance generalisation, interpretability, and controllability. While current Transformer-based language models have shown strong performance on Natural Language Inference (NLI) tasks, they often rely on memorisation rather than explicit rule-based generalisation. This work investigates how human-interpretable reasoning rules can be explicitly encoded within LMs with the support of Language Variational Autoencoders (VAEs), as a mechanism for generative control. We propose a complete pipeline for learning reasoning rules within Transformer-based language VAEs. This pipeline encompasses three rule-based reasoning tasks, a supporting theoretical framework, and a practical end-to-end architecture. The experiment illustrates the following findings: Disentangled reasoning: Under explicit signal supervision, reasoning rules (viewed as functional mappings) can be disentangled within the encoder’s parametric space. This separation results in distinct clustering of rules in the output feature space. Prior knowledge injection: injecting rule-based constraints into the Query enables the model to more effectively retrieve the stored Value from memory based on Key. This approach offers a simple method for integrating prior knowledge into decoder-only language models. Moreover, we found that FFN layers are better than attention layers at preserving the separation of reasoning rules in the model's parameters.

Ruijie Zhang, Bixin Zeng, Shengpeng Wang, Fuhui Zhou, Wei Wang

Millimeter-wave radar offers a promising sensing modality for autonomous systems thanks to its robustness in adverse conditions and low cost. However, its utility is significantly limited by the sparsity and low resolution of radar point clouds, which poses challenges for tasks requiring dense and accurate 3D perception. Despite that recent efforts have shown great potential by exploring generative approaches to address this issue, they often rely on dense voxel representations that are inefficient and struggle to preserve structural detail. To fill this gap, we make the key observation that latent diffusion models (LDMs), though successful in other modalities, have not been effectively leveraged for radar-based 3D generation due to a lack of compatible representations and conditioning strategies. We introduce RaLD, a framework that bridges this gap by integrating scene-level frustum-based LiDAR autoencoding, order-invariant latent representations, and direct radar spectrum conditioning. These insights lead to a more compact and expressive generation process. Experiments show that RaLD produces dense and accurate 3D point clouds from raw radar spectrums, offering a promising solution for robust perception in challenging environments.

Shilin Zhang, Suixue Wang, Qingchen Zhang, Xiulong Liu

Recent advances in spatial transcriptomics have enabled the integration of gene expression profiles with precise spatial coordinates, which have facilitated the exploration of tumor occurrence and development mechanisms, as well as the development of more effective targeted and immunotherapy approaches for tumor treatment. Deciphering cell type represents a critical challenge in spatial transcriptomics research. Existing methods are limited by the pervasive “dropout” events in spatial transcriptomics, hindering their ability to fully capture the relationship between spatial location and gene expression, thereby compromising the performance of cell type deconvolution. To address these limitations, we propose a spatial-aware masked graph transformer-diffusion model (SAMGTD) for enhanced cell type deconvolution in spatial transcriptomics. For spatial transcriptomics, the masked graph transformer model is designed to adaptively capture complex dependencies between spatial locations and gene expression. It employs a masking strategy that guides the model to focus on important local information during training, while the multi-head attention mechanism captures global context. More importantly, the spatial diffusion model is constructed to achieve the dual enhancement of spatial transcriptomics, including denoising and data imputation. It incorporates the multi-head attention mechanism and residual blocks, effectively addressing the “dropout” issue commonly encountered in spatial transcriptomics. For scRNA-seq, we construct a variational autoencoder to reduce noise interference while preserving key gene expression information. Finally, we construct a spatial-aware contrastive learning model to integrate scRNA-seq and spatial transcriptomics for cell type deconvolution. Experiments conducted on three datasets demonstrate that SAMGTD outperforms baseline methods.

Haoyu Wang, Xiaozhe Xin, Xiaoyu Qin, Meiguang Jin, Junfeng Ma, Dan Xu, Jia Jia

Recent advances in audio-driven talking-head synthesis have brought lip-sync precision close to human perception, yet emotional fidelity and real-time inference remain open challenges. Existing pipelines typically disentangle lip articulation, facial expression, and head pose in latent space; this rigid factorization ignores the intrinsic coupling between articulation and affect — e.g., downward lip corners when sad—thus limiting expressiveness. We cast speech-conditioned facial motion as a sample from an emotion-conditioned distribution in a motion latent space. Concretely, we (i) learn a motion dictionary of orthogonal bases with an autoencoder via self-supervision, (ii) construct emotion-conditioned sub-spaces within the latent space, and (iii) design a layer-progressive cross-attention fusion module that modulates a flow-matching sampler with both audio and emotion signals. Only ten reverse ODE steps are required to generate a motion-latent trajectory, enabling real-time end-to-end latency. Extensive experiments on MEAD and RAVDESS show that our method outperforms recent GAN- and diffusion-based baselines in emotion accuracy while running at around 75 FPS on a single desktop GPU. The proposed framework delivers the first emotionally expressive Audio2Face system that simultaneously achieves lip-sync accuracy, affective realism, and real-time performance.

Yuting Tang, Weibang Jiang, Shanglin Li, Yong Li, Chenyu Liu, Xinliang Zhou, Yi Ding, Cuntai Guan

Large-scale EEG foundation models have shown strong generalization across a range of downstream tasks, but their training remains resource-intensive due to the volume and variable quality of EEG data. In this work, we introduce EEG-DLite, a data distillation framework that enables more efficient pre-training by selectively removing noisy and redundant samples from large EEG datasets. EEG-DLite begins by encoding EEG segments into compact latent representations using a self-supervised autoencoder, allowing sample selection to be performed efficiently and with reduced sensitivity to noise. Based on these representations, EEG-DLite filters out outliers and minimizes redundancy, resulting in a smaller yet informative subset that retains the diversity essential for effective foundation model training. Through extensive experiments, we demonstrate that training on only 5 percent of a 2,500-hour dataset curated with EEG-DLite yields performance comparable to, and in some cases better than, training on the full dataset across multiple downstream tasks. To our knowledge, this is the first systematic study of pre-training data distillation in the context of EEG foundation models. EEG-DLite provides a scalable and practical path toward more effective and efficient physiological foundation modeling.

Feng Cao, Zishuo Feng, Jicong Zhang, Wei Shi

Extracellular recordings are transient voltage fluctuations in the vicinity of neurons, serving as a fundamental modality in neuroscience for decoding brain activity at single-neuron resolution. Spike sorting, the process of attributing each detected spike to its corresponding neuron, is a pivotal step in brain sensing pipelines. However, it remains challenging under low signal-to-noise ratio (SNR), electrode drift, and cross-session variability. In this paper, we propose HuiduRep, a robust self-supervised representation learning framework that extracts discriminative and generalizable features from extracellular recordings. By integrating contrastive learning with a denoising autoencoder, HuiduRep learns latent representations robust to noise and drift. With HuiduRep, we develop a spike sorting pipeline that clusters spike representations without ground truth labels. Experiments on hybrid and real-world datasets demonstrate that HuiduRep achieves strong robustness. Furthermore, the pipeline significantly outperforms state-of-the-art tools such as KiloSort4 and MountainSort5 on accuracy and precision on diverse datasets. These findings demonstrate the potential of self-supervised spike representation learning as a foundational tool for robust and generalizable processing of extracellular recordings.

Haochen Yuan, Yang Zhang, Xiang He, Quan Z. Sheng, Zhongjie Wang

With the rise of cloud-edge collaboration, recommendation services are increasingly trained in distributed environments. Federated Recommendation (FR) enables such multi-end collaborative training while preserving privacy by sharing model parameters instead of raw data. However, the large number of parameters, primarily due to the massive item embeddings, significantly hampers communication efficiency. While existing studies mainly focus on improving the efficiency of FR models, they largely overlook the issue of embedding parameter overhead. To address this gap, we propose a FR training framework with Parameter-Efficient Fine-Tuning (PEFT) based embedding designed to reduce the volume of embedding parameters that need to be transmitted. Our approach offers a lightweight, plugin-style solution that can be seamlessly integrated into existing FR methods. In addition to incorporating common PEFT techniques such as LoRA and Hash-based encoding, we explore the use of Residual Quantized Variational Autoencoders (RQ-VAE) as a novel PEFT strategy within our framework. Extensive experiments across various FR model backbones and datasets demonstrate that our framework significantly reduces communication overhead while improving accuracy.