Author name cluster

Wen Wen

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

10 papers

2 author rows

AAAI Conference 2026 Conference Paper

Beyond Sharpness: A Flatness Decomposition Framework for Efficient Continual Learning

Yanan Chen
Tieliang Gong
Yunjiao Zhang
Wen Wen

Continual Learning (CL) aims to enable models to sequentially learn multiple tasks without forgetting previous knowledge. Recent studies have shown that optimizing towards flatter loss minima can improve model generalization. However, existing sharpness-aware methods for CL suffer from two key limitations: (1) they treat sharpness regularization as a unified signal without distinguishing the contributions of its components. and (2) they introduce substantial computational overhead that impedes practical deployment. To address these challenges, we propose FLAD, a novel optimization framework that decomposes sharpness-aware perturbations into gradient-aligned and stochastic-noise components, and show that retaining only the noise component promotes generalization. We further introduce a lightweight scheduling scheme that enables FLAD to maintain significant performance gains even under constrained training time. FLAD can be seamlessly integrated into various CL paradigms and consistently outperforms standard and sharpness-aware optimizers in diverse experimental settings, demonstrating its effectiveness and practicality in CL.

PDF Details DOI

AAAI Conference 2026 Conference Paper

Recovering Coherent Affective Patterns: Addressing Modality Missing in Multimodal Sentiment Analysis

Huiting Huang
Tieliang Gong
Kai He
Wen Wen
Weizhan Zhang
Mengling Feng

Multimodal sentiment analysis (MSA) seeks to decode human emotions by integrating heterogeneous modalities. However, real-world scenarios often involve missing or misaligned data due to sensor failures or transmission errors, leading to disrupted temporal dynamics and degraded cross-modal correlations. To address these challenges, we propose RECAP (REcovery of Coherent Affective Patterns), a robust two-stage framework to restore temporal and structural emotional integrity under modality incompleteness. The first stage employs a causality-aware adversarial generator for multi-granularity temporal reconstruction, complemented by a contrastive mutual information factorization module that disentangles shared and modality-specific semantics. The second stage introduces a mutual information-guided attention fusion mechanism with a ranking-based objective, enabling adaptive integration of complementary signals for refined prediction. Extensive experiments on MOSI, MOSEI, and SIMS under various missing-modality conditions demonstrate that RECAP consistently outperforms state-of-the-art methods. Notably, it improves ACC-7 on MOSI by 2.71 percentage points and F1 on SIMS by 6.38 percentage points. These results verify the performance of RECAP in terms of capturing fine-grained emotional cues and robustness.

PDF Details DOI

TIST Journal 2025 Journal Article

Modeling Multi-Seasonal Multi-Behavior Dependency for Temporal Recommendation

Shichao Liang
Wen Wen
Yali Feng
Ruichu Cai
Zhifeng Hao

Mining temporal patterns from user behaviors has long been investigated, but most of the existing work centers on single-type user–item interactions, such as purchase or click, which fails to take advantage of the user’s diversified interests revealed by various types of behavior. However, capturing patterns from different behavior sequences and modeling the complex inter-correlation between them are non-trivial tasks, as the high sparsity of type-related interactions, multi-seasonality of individual behaviors, and time-variant dependency of multi-type activities make it really challenging. To address these challenges, we propose a novel framework that aims to model the M ulti-Seasonal M ulti-Behavior Dep endencies (MMDep) both within and across the multi-type behavior sequences. In the proposed model, an item co-occurrence matrix factorization strategy is introduced to alleviate the sparsity issue in type-related behavior sequences. And a temporal dependency module that incorporates multi-scale EMA mechanism is utilized to capture the multi-seasonal dependencies within individual sequences. Moreover, a cross-behavior dependency module is employed to learn the time-variant dependency among different behaviors. Extensive experiments on three real-world datasets demonstrate that the proposed MMDep performs significantly better than the state-of-the-art baselines. And it may provide some new insights and tools on how to leverage multi-behavior data for better temporal recommendation.

Details DOI

IJCAI Conference 2024 Conference Paper

Towards Sharper Generalization Bounds for Adversarial Contrastive Learning

Wen Wen
Han Li
Tieliang Gong
Hong Chen

Recently, the enhancement on the adversarial robustness of machine learning algorithms has gained significant attention across various application domains. Given the widespread label scarcity issue in real-world data, adversarial contrastive learning (ACL) has been proposed to adversarially train robust models using unlabeled data. Despite the empirical success, its generalization behavior remains poorly understood and far from being well-characterized. This paper aims to address this issue from a learning theory perspective. We establish novel high-probability generalization bounds for the general Lipschitz loss functions. The derived bounds scale O(log(k)) with respect to the number of negative samples k, which improves the existing linear dependency bounds. Our results are generally applicable to many prediction models, including linear models and deep neural networks. In particular, we obtain an optimistic generalization bound O(1/n) under the smoothness assumption of the loss function on the sample size n. To the best of our knowledge, this is the first fast-rate bound valid for ACL. Empirical evaluations on real-world datasets verify our theoretical findings.

PDF Details DOI

IJCAI Conference 2023 Conference Paper

Generalization Bounds for Adversarial Metric Learning

Wen Wen
Han Li
Hong Chen
Rui Wu
Lingjuan Wu
Liangxuan Zhu

Recently, adversarial metric learning has been proposed to enhance the robustness of the learned distance metric against adversarial perturbations. Despite rapid progress in validating its effectiveness empirically, theoretical guarantees on adversarial robustness and generalization are far less understood. To fill this gap, this paper focuses on unveiling the generalization properties of adversarial metric learning by developing the uniform convergence analysis techniques. Based on the capacity estimation of covering numbers, we establish the first high-probability generalization bounds with order O(n^{-1/2}) for adversarial metric learning with pairwise perturbations and general losses, where n is the number of training samples. Moreover, we obtain the refined generalization bounds with order O(n^{-1}) for the smooth loss by using local Rademacher complexity, which is faster than the previous result of adversarial pairwise learning, e. g. , adversarial bipartite ranking. Experimental evaluation on real-world datasets validates our theoretical findings.

PDF Details DOI

EAAI Journal 2020 Journal Article

Robust RGB-D tracking via compact CNN features

Yong Wang
Xian Wei
Lingkun Luo
Wen Wen
Yang Wang

Feature representation is at the core of visual tracking. This paper presents a robust tracking method in RGB-D videos. Firstly, the RGB and depth images are separately encoded using a hierarchical convolutional neural network (CNN) features. Secondly, in order to reduce computation cost, we exploit random projection to compress the CNN features. The high dimensional CNN features are randomly projected into a low dimensional feature space. The correlation filter tracking framework is then independently carried out in RGB and depth images. And backward tracking scheme is adopted to evaluate the tracking results in these two images. The final position is determined according to the tracked location in the two image channels. In addition, model updating is implemented adaptively. Our tracker is evaluated on two RGB-D benchmark datasets and achieves comparable results to the other state-of-the-art RGB-D tracking methods.

Details DOI

YNICL Journal 2020 Journal Article

Theta oscillations in prolactinomas: Neurocognitive deficits in executive controls

Chenglong Cao
Wen Wen
Binbin Liu
Pan Ma
Sheng Li
Guozheng Xu
Jian Song

Impairment of cognitive functions has been reported in prolactinomas. However, the electrophysiological mechanisms of response activation and response inhibition in prolactinomas remain unclear. We recorded participants' scalp electroencephalography (EEG) in a visual Go/Nogo task. Compared to the healthy controls (HCs), the patients demonstrated worse performance and their prolactin (PRL) levels negatively correlated with behavioral results. Meanwhile, patients' P300 amplitudes in the Go and Nogo conditions were smaller than the HCs. The amplitudes of N200nogo in patients were smaller than the HCs as well. Lower frontal theta power was found in the patients than the HCs in both Go and Nogo conditions, which indicated a deficit in response activation and inhibition. Moreover, the PRL levels mediated the relationship between frontal theta power and behavior performance, implying that lower frontal theta power caused the dysfunction of response control by abnormally high PRL levels. Patients also showed lower occipital alpha power than the HCs, which suggested that the impaired response inhibition may arise from deficient attention control. Taken together, the present study revealed the neurocognitive discrepancies between prolactinomas and the HCs. The frontal theta oscillation was highlighted as the electrophysiological markers of the impaired response control in prolactinomas.

Details DOI

YNIMG Journal 2018 Journal Article

Enhanced perceptual processing of self-generated motion: Evidence from steady-state visual evoked potentials

Wen Wen
Elisa Brann
Steven Di Costa
Patrick Haggard

The sense of agency emerges when our voluntary actions produce anticipated or predictable outcomes in the external world. It remains unclear how the sense of control also influences our perception of the external world. The present study examined perceptual processing of self-generated motion versus non-self-generated motion using steady-state visual evoked potentials (SSVEPs). Participants continuously moved their finger on a touchpad to trigger the movements of two shapes (Experiment 1) or two groups of dots (Experiment 2) on a monitor. Degree of control was manipulated by varying the spatial relation between finger movement and stimulus trajectory across conditions. However, the velocity, onset time, and offset time of visual stimuli always corresponded to participants' finger movement. Stimuli flickered at a frequency of either 7. 5 Hz or 10 Hz, thus SSVEPs of these frequencies and their harmonics provided a frequency-tagged measurement of perceptual processing. Participants triggered the motion of all stimuli simultaneously, but had greater levels of control over some stimuli than over others. Their task was to detect a brief colour change on the border(s) of one shape (Experiment 1) or of one group of dots (Experiment 2). Although control over shapes/dots was irrelevant to the visual detection task, we found stronger SSVEPs for stimuli that were under a high level of control, compared with the stimuli that were under a low level of control. Our results suggest that the spatial regularity between self-generated movements and visual input boosted the neural responses underlying perceptual processing. Our results support the preactivation account of sensory attenuation, suggesting that perceptual processing of self-generated events is enhanced rather than inhibited.

Details DOI

UAI Conference 2018 Conference Paper

Unsupervised Multi-view Nonlinear Graph Embedding

Jiaming Huang
Zhao Li 0007
Vincent W. Zheng
Wen Wen
Yifan Yang 0001
Yuanmi Chen

ei, 1 xi, 1 In this paper, we study the unsupervised multi-view graph embedding (UMGE) problem, which aims to learn graph embedding from multiple perspectives in an unsupervised manner. However, the vast majority of multiview learning work focuses on non-graph data, and surprisingly there are limited work on UMGE. By systematically analyzing different existing methods for UMGE, we discover that cross-view and nonlinearity play a vital role in efficiently improving graph embedding quality. Motivated by this concept, we develop an unsupervised Multi-viEw nonlineaR Graph Embedding (MERGE) approach to model relational multi-view consistency. Experimental results on five benchmark datasets demonstrate that MERGE significantly outperforms the state-of-the-art baselines in terms of accuracy in node classification tasks without sacrificing the computational efficiency.

Details

YNIMG Journal 2015 Journal Article

Layer-specific response properties of the human lateral geniculate nucleus and superior colliculus

Peng Zhang
Hao Zhou
Wen Wen
Sheng He

The human LGN and SC consist of distinct layers, but their layer-specific response properties remain poorly understood. In this fMRI study, we characterized visual response properties of the magnocellular (M) and parvocellular (P) layers of the human LGN, as well as at different depths in the SC. Results show that fMRI is capable of resolving layer-specific signals from the LGN and SC. Compared to the P layers of the LGN, the M layers preferred higher temporal frequency, lower spatial frequency stimuli, and their responses saturated at lower contrast. Furthermore, the M layers are colorblind while the P layers showed robust response to both chromatic and achromatic stimuli. Visual responses in the SC were strongest in the superficial voxels, which showed similar spatiotemporal and contrast response properties as the M layers of the LGN, but were sensitive to color and responded strongly to isoluminant color stimulus. Thus, the non-invasive fMRI measures show that the M and P layers of human LGN have similar response properties as that observed in non-human primates and the superficial layers of the human SC prefer transient inputs but are not colorblind.

Details DOI