Author name cluster
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.
EAAI Journal 2026 Journal Article
EAAI Journal 2026 Journal Article
With the rapid development of online social networks, identifying the most influential user groups and designating them as source nodes for information dissemination has attracted increasing attention, as this approach can help maximize information dissemination efficiency. In particular, Influence Maximization (IM) in multi-entity social networks becomes a very hot topic recently. However, existing methodologies confront two critical limitations: first, most studies fail to adequately account for the fact that the correlations between the target item and other items can change with users’ evolving interests, thus oversimplifying the complex relationships between items. Second, an unresolved trade-off between computational efficiency and algorithmic precision, thereby restricting their scalability in large-scale network applications. To address these challenges, this paper proposes a Long-term and Short-term interest Fusion-based Reverse Influence Sampling model (named LSF-RIS) for the issue of multi-entity IM. LSF-RIS consists of two main components: (1) a Deep Long-term and Short-term interest Fusion (DLSF) module that dynamically models user preferences through temporal interest integration, thereby enhancing the prediction accuracy of user activation probabilities for target items; and (2) a Weighted Core Decomposition-enhanced Reverse Influence Sampling (RIS-WCD) mechanism, which optimizes the network topology by pruning non-critical nodes, retains the core nodes and crucial connections within the network, and thereby achieves enhanced computational efficiency without compromising sampling accuracy. Extensive experiments on three public social network datasets demonstrate the superiority of LSF-RIS over state-of-the-art methods, as evidenced by metrics of both accuracy and computational efficiency.
AAAI Conference 2025 Conference Paper
Dense video captioning (DVC) aims to describe multiple events within a video, and its performance is greatly affected by the accuracy of video event detection. Video event detection involves predicting the proposal boundaries (start and end times) and the classification score of each event in a video. Recently, a few methods have applied diffusion models originally designed for image object detection to detect events in DVC. These methods add noise to the ground-truth event proposal boundaries, and subsequently learn the denoising process. However, these methods often overlook the fundamental differences between videos and images. We observe that, whereas in images the important information for object classification is normally around the boundaries of the ground-truth boxes, in videos the key information for event classification is typically centered in the middle of ground-truth event proposals. As a result, the classification module in these existing diffusion models becomes insensitive to boundary changes introduced by the added noise, leading to sub-optimal performance. This paper introduces DiffDVC, an innovative diffusion model for DVC. The core of DiffDVC is a boundary-sensitive detector. The detector increases the sensitivity of the classification module to boundary changes by focusing on frames within a specific range around the start and end times of noisy event proposals. Additionally, this range is dynamically adjusted to suit different event proposals. Comprehensive experiments on ActivityNet-1.3, ActivityNet Captions, and YouCook2 datasets show DiffDVC achieving superior performance.
NeurIPS Conference 2025 Conference Paper
Reinforcement learning (RL) has become an effective approach for fine-tuning large language models (LLMs), particularly to enhance their reasoning capabilities. However, RL fine-tuning remains highly resource-intensive, and existing work has largely overlooked the problem of data efficiency. In this paper, we propose two techniques to improve data efficiency in LLM RL fine-tuning: difficulty-targeted online data selection and rollout replay. We introduce the notion of adaptive difficulty to guide online data selection, prioritizing questions of moderate difficulty that are more likely to yield informative learning signals. To estimate adaptive difficulty efficiently, we develop an attention-based framework that requires rollouts for only a small reference set of questions. The adaptive difficulty of the remaining questions is then estimated based on their similarity to this set. To further reduce rollout cost, we introduce a rollout replay mechanism inspired by experience replay in traditional RL. This technique reuses recent rollouts, lowering per-step computation while maintaining stable updates. Experiments across 6 LLM-dataset combinations show that our method reduces RL fine-tuning time by 23% to 62% while reaching the same level of performance as the original GRPO algorithm. Our code repository is available at https: //github. com/ASTRAL-Group/data-efficient-llm-rl/.
NeurIPS Conference 2024 Conference Paper
Offline reinforcement learning (RL) leverages pre-collected datasets to train optimal policies. Diffusion Q-Learning (DQL), introducing diffusion models as a powerful and expressive policy class, significantly boosts the performance of offline RL. However, its reliance on iterative denoising sampling to generate actions slows down both training and inference. While several recent attempts have tried to accelerate diffusion-QL, the improvement in training and/or inference speed often results in degraded performance. In this paper, we introduce a dual policy approach, Diffusion Trusted Q-Learning (DTQL), which comprises a diffusion policy for pure behavior cloning and a practical one-step policy. We bridge the two polices by a newly introduced diffusion trust region loss. The diffusion policy maintains expressiveness, while the trust region loss directs the one-step policy to explore freely and seek modes within the region defined by the diffusion policy. DTQL eliminates the need for iterative denoising sampling during both training and inference, making it remarkably computationally efficient. We evaluate its effectiveness and algorithmic characteristics against popular Kullback-Leibler (KL) based distillation methods in 2D bandit scenarios and gym tasks. We then show that DTQL could not only outperform other methods on the majority of the D4RL benchmark tasks but also demonstrate efficiency in training and inference speeds. The PyTorch implementation is available at https: //github. com/TianyuCodings/Diffusion Trusted Q_Learning.
TIST Journal 2024 Journal Article
Understanding and accurately predicting cellular traffic data is vital for communication operators and device users, as it facilitates efficient resource allocation and ensures superior service quality. However, large-scale cellular traffic data forecasting remains challenging due to intricate temporal variations and complex spatial relationships. This article proposes a Knowledge Graph Driven Decomposition Approach (KGDA) for precise cellular traffic prediction. The KGDA breaks down the impact of static environmental factors and dynamic autocorrelations of cellular traffic time series, enabling the capture of overall traffic changes and understanding of traffic dependence on past values. Specifically, we propose an urban knowledge graph to capture the static environmental context of base stations, mapping these entities into the same latent space while retaining static environmental knowledge. The cellular traffic is divided into a regular pattern and fluctuating residual components, with the KGDA comprising four modules: a Knowledge Graph Representation Learning model, a traffic regular pattern prediction module, a traffic residual dynamic prediction module, and an attentional fusion module. The first leverages graph neural networks to extract spatial contexts and predict regular patterns, the second utilizes the Bi-directional Long Short-Term Memory (Bi-LSTM) model to capture autocorrelations of traffic time series, and the final module integrates the patterns and residuals to produce the final prediction result. Comprehensive experiments demonstrate that our proposed model outperforms state-of-the-art models by more than 10% in forecasting cellular traffic.
NeurIPS Conference 2023 Conference Paper
We introduce beta diffusion, a novel generative modeling method that integrates demasking and denoising to generate data within bounded ranges. Using scaled and shifted beta distributions, beta diffusion utilizes multiplicative transitions over time to create both forward and reverse diffusion processes, maintaining beta distributions in both the forward marginals and the reverse conditionals, given the data at any point in time. Unlike traditional diffusion-based generative models relying on additive Gaussian noise and reweighted evidence lower bounds (ELBOs), beta diffusion is multiplicative and optimized with KL-divergence upper bounds (KLUBs) derived from the convexity of the KL divergence. We demonstrate that the proposed KLUBs are more effective for optimizing beta diffusion compared to negative ELBOs, which can also be derived as the KLUBs of the same KL divergence with its two arguments swapped. The loss function of beta diffusion, expressed in terms of Bregman divergence, further supports the efficacy of KLUBs for optimization. Experimental results on both synthetic data and natural images demonstrate the unique capabilities of beta diffusion in generative modeling of range-bounded data and validate the effectiveness of KLUBs in optimizing diffusion models, thereby making them valuable additions to the family of diffusion-based generative models and the optimization techniques used to train them.
NeurIPS Conference 2023 Conference Paper
We present Prompt Diffusion, a framework for enabling in-context learning in diffusion-based generative models. Given a pair of task-specific example images, such as depth from/to image and scribble from/to image, and a text guidance, our model automatically understands the underlying task and performs the same task on a new query image following the text guidance. To achieve this, we propose a vision-language prompt that can model a wide range of vision-language tasks and a diffusion model that takes it as input. The diffusion model is trained jointly on six different tasks using these prompts. The resulting Prompt Diffusion model becomes the first diffusion-based vision-language foundation model capable of in-context learning. It demonstrates high-quality in-context generation for the trained tasks and effectively generalizes to new, unseen vision tasks using their respective prompts. Our model also shows compelling text-guided image editing results. Our framework aims to facilitate research into in-context learning for computer vision. We share our code and pre-trained models at https: //github. com/Zhendong-Wang/Prompt-Diffusion.
NeurIPS Conference 2023 Conference Paper
Diffusion models are powerful, but they require a lot of time and data to train. We propose Patch Diffusion, a generic patch-wise training framework, to significantly reduce the training time costs while improving data efficiency, which thus helps democratize diffusion model training to broader users. At the core of our innovations is a new conditional score function at the patch level, where the patch location in the original image is included as additional coordinate channels, while the patch size is randomized and diversified throughout training to encode the cross-region dependency at multiple scales. Sampling with our method is as easy as in the original diffusion model. Through Patch Diffusion, we could achieve $\mathbf{\ge 2\times}$ faster training, while maintaining comparable or better generation quality. Patch Diffusion meanwhile improves the performance of diffusion models trained on relatively small datasets, $e. g. $, as few as 5, 000 images to train from scratch. We achieve outstanding FID scores in line with state-of-the-art benchmarks: 1. 77 on CelebA-64$\times$64, 1. 93 on AFHQv2-Wild-64$\times$64, and 2. 72 on ImageNet-256$\times$256. We share our code and pre-trained models at https: //github. com/Zhendong-Wang/Patch-Diffusion.
AIIM Journal 2023 Journal Article
In recent years, machine learning methods have been rapidly adopted in the medical domain. However, current state-of-the-art medical mining methods usually produce opaque, black-box models. To address the lack of model transparency, substantial attention has been given to developing interpretable machine learning models. In the medical domain, counterfactuals can provide example-based explanations for predictions, and show practitioners the modifications required to change a prediction from an undesired to a desired state. In this paper, we propose a counterfactual solution MedSeqCF for preventing the mortality of three cohorts of ICU patients, by representing their electronic health records as medical event sequences, and generating counterfactuals by adopting and employing a text style-transfer technique. We propose three model augmentations for MedSeqCF to integrate additional medical knowledge for generating more trustworthy counterfactuals. Experimental results on the MIMIC-III dataset strongly suggest that augmented style-transfer methods can be effectively adapted for the problem of counterfactual explanations in healthcare applications and can further improve the model performance in terms of validity, BLEU-4, local outlier factor, and edit distance. In addition, our qualitative analysis of the results by consultation with medical experts suggests that our style-transfer solutions can generate clinically relevant and actionable counterfactual explanations.
NeurIPS Conference 2020 Conference Paper
To improve the sample efficiency of policy-gradient based reinforcement learning algorithms, we propose implicit distributional actor-critic (IDAC) that consists of a distributional critic, built on two deep generator networks (DGNs), and a semi-implicit actor (SIA), powered by a flexible policy distribution. We adopt a distributional perspective on the discounted cumulative return and model it with a state-action-dependent implicit distribution, which is approximated by the DGNs that take state-action pairs and random noises as their input. Moreover, we use the SIA to provide a semi-implicit policy distribution, which mixes the policy parameters with a reparameterizable distribution that is not constrained by an analytic density function. In this way, the policy's marginal distribution is implicit, providing the potential to model complex properties such as covariance structure and skewness, but its parameter and entropy can still be estimated. We incorporate these features with an off-policy algorithm framework to solve problems with continuous action space and compare IDAC with state-of-the-art algorithms on representative OpenAI Gym environments. We observe that IDAC outperforms these baselines in most tasks. Python code is provided.