Jin Dong

AAAI Conference 2026 Conference Paper

FedSEA-LLaMA: A Secure, Efficient and Adaptive Federated Splitting Framework for Large Language Models

• Zishuai Zhang
• Hainan Zhang
• Weihua Li
• Qinnan Zhang
• Jin Dong
• Yongxin Tong
• Zhiming Zheng

Private data holds promise for improving LLMs due to its high quality, but its scattered distribution across data silos and the high computational demands of LLMs limit their deployment in federated environments. To address this, the transformer-based federated split models are proposed, which offload most model parameters to the server (or distributed clients) while retaining only a small portion on the client to ensure data privacy. Despite this design, they still face three challenges: 1) Peer-to-peer key encryption struggles to secure transmitted vectors effectively; 2) The auto-regressive nature of LLMs means that federated split learning can only train and infer sequentially, causing high communication overhead; 3) Fixed partition points lack adaptability to downstream tasks. In this paper, we introduce FedSEA-LLaMA, a Secure, Efficient, and Adaptive Federated splitting framework based on LLaMA2. First, we inject Gaussian noise into forward-pass hidden states to enable secure end-to-end vector transmission. Second, we employ attention-mask compression and KV cache collaboration to reduce communication costs, accelerating training and inference. Third, we allow users to dynamically adjust the partition points for input/output blocks based on specific task requirements. Experiments on natural language understanding, summarization, and conversational QA tasks show that FedSEA-LLaMA maintains performance comparable to centralized LLaMA2 and achieves up to 8× speedups in training and inference. Further analysis of privacy attacks and different partition points also demonstrates the effectiveness of FedSEA-LLaMA in security and adaptability.

JBHI Journal 2023 Journal Article

SwinDAE: Electrocardiogram Quality Assessment Using 1D Swin Transformer and Denoising AutoEncoder

• Guanyu Chen
• Tianyi Shi
• Baoxing Xie
• Zhicheng Zhao
• Zhu Meng
• Yadong Huang
• Jin Dong

Objective: Electrocardiogram (ECG) signals have wide-ranging applications in various fields, and thus it is crucial to identify clean ECG signals under different sensors and collection scenarios. Despite the availability of a variety of deep learning algorithms for ECG quality assessment, these methods still lack generalization across different datasets, hindering their widespread use. Methods: In this paper, an effective model named Swin Denoising AutoEncoder (SwinDAE) is proposed. Specifically, SwinDAE uses a DAE as the basic architecture, and incorporates a 1D Swin Transformer during the feature learning stage of the encoder and decoder. SwinDAE was first pre-trained on the public PTB-XL dataset after data augmentation, with the supervision of signal reconstruction loss and quality assessment loss. Specially, the waveform component localization loss is proposed in this paper and used for joint supervision, guiding the model to learn key information of signals. The model was then fine-tuned on the finely annotated BUT QDB dataset for quality assessment. Results: SwinDAE achieved 0. 02–0. 13 mean F1 score improvement on the BUT QDB dataset compared to multiple deep learning methods, and demonstrated applicability on two other datasets. Conclusion: The proposed SwinDAE shows strong generalization ability on different datasets, and surpasses other state-of-the-art deep learning methods on multiple evaluation metrics. In addition, the statistical analysis for SwinDAE prove the significance of the performance and the rationality of the prediction. Significance: SwinDAE can learn the commonality between high-quality ECG signals, exhibiting excellent performance in the application of cross-sensors and cross-collection scenarios.

IJCAI Conference 2009 Conference Paper

• Xinxin Bai
• Gang Chen
• Qiming Tian
• Wenjun Yin
• Jin Dong

Location plays a very important role in the retail business due to its huge and long-term investment. In this paper, we propose a novel semisupervised regression model for evaluating convenience store location based on spatial data analysis. First, the input features for each convenience store can be extracted by analyzing the elements around it based on a geographic information system, and the turnover is used to evaluate its performance. Second, considering the practical application scenario, a manifold regularization model with one semi-supervised performance information constraint is provided. The promising experimental results in the real-world dataset demonstrate the effectiveness of the proposed approach in performance prediction of certain candidate locations for new convenience store opening.

IJCAI Conference 2009 Conference Paper

• Fei Wang
• Bin Zhang
• Ta-Hsin Li
• Wen Jun Yin
• Jin Dong
• Tao Li

In this paper, we consider a general problem of semi-supervised preference learning, in which we assume that we have the information of the extreme cases and some ordered constraints, our goal is to learn the unknown preferences of the other places. Taking the potential housing place selection problem as an example, we have many candidate places together with their associated information (e. g. , position, environment), and we know some extreme examples (i. e. several places are perfect for building a house, and several places are the worst that cannot build a house there), and we know some partially ordered constraints (i. e. for two places, which place is better), then how can we judge the preference of one potential place whose preference is unknown beforehand? We propose a Bayesian framework based on Gaussian process to tackle this problem, from which we not only solve for the unknown preferences, but also the hyperparameters contained in our model.