Institution profile

Kyung Hee University

Academic institutionasia · kr
Official website
Research library346linked papers
Opportunities0open roles
Selected work

Representative Papers

MAD: Multi-Alignment MEG-to-Text Decoding

Jun 03, 2024arXiv.org

Current non-invasive BCI-based language decoding faces three key bottlenecks: underutilization of magnetoencephalography (MEG) signals, poor cross-sentence generalization, and absence of multimodal fusion. Method: We propose the first end-to-end, multi-aligned MEG-to-text framework for natural language reconstruction from entirely unseen sentences. Our approach introduces a Transformer-based architecture that jointly aligns neural time series, phonemes, and semantics, integrating self-supervised pretraining with cross-modal contrastive learning to systematically unify speech, semantic, and dynamic temporal information. Results: On the Gwilliams dataset, our method achieves a BLEU-1 score of 10.44—improving by 4.95 (+93%) over the strongest baseline—demonstrating substantially enhanced open-vocabulary text generation capability. This work breaks critical limitations in generalizability and multimodal integration for non-invasive brain–computer interface–based language reconstruction.

18 citations3 influentialRead paper

SpectrumFM: A Foundation Model for Intelligent Spectrum Management

May 02, 2025arXiv.org

To address the low recognition accuracy, slow convergence, and poor generalization of existing small-scale models in dynamic spectrum environments, this paper proposes SpectrumFM—a spectral foundation model. Methodologically, SpectrumFM integrates CNNs with multi-head self-attention to enhance IQ-signal representation learning; introduces the first foundation-model paradigm for spectrum analysis, featuring dual self-supervised pretraining tasks—masked signal reconstruction and next-time-step signal prediction; and employs parameter-efficient fine-tuning (e.g., LoRA) for cross-task transfer. Experiments demonstrate significant improvements: 12.1% higher accuracy in automatic modulation classification (AMC), 9.3% gain in wireless technology classification (WTC), an AUC of 0.97 for spectrum sensing at −4 dB SNR, over 10% improvement in anomaly detection performance, faster convergence, and markedly enhanced few-shot adaptation capability.

3 citations1 influentialRead paper

MQ-GNN: A Multi-Queue Pipelined Architecture for Scalable and Efficient GNN Training

Jan 08, 2026IEEE Access

This work addresses the inefficiencies in graph neural network (GNN) training caused by inefficient minibatch generation, data transfer bottlenecks, and synchronization overhead across multiple GPUs, which collectively lead to low hardware utilization. To overcome these challenges, the authors propose MQ-GNN, a framework featuring a multi-queue pipelined architecture that interleaves different training stages. MQ-GNN introduces the Ready-to-Update asynchronous consistency model (RaCoM) to enable asynchronous gradient sharing with adaptive periodic synchronization. It further incorporates a global neighbor sampling cache and an adaptive queue sizing strategy to enhance throughput while preserving model consistency. Experimental results on four large-scale datasets demonstrate that MQ-GNN achieves up to 4.6× speedup over ten baseline models, improves GPU utilization by 30%, and maintains competitive accuracy.

2 citationsRead paper

Simultaneously Transmitting and Reflecting Surfaces (STARS) for Multi-Functional 6G

Jan 01, 2025IEEE Network

Emerging 6G multimodal networks demand seamless integration of communication, sensing, computing, and caching—challenging conventional reconfigurable intelligent surfaces (RISs), which support only unidirectional communication. Method: This paper proposes simultaneous transmitting-and-reflecting intelligent surfaces (STARS), capable of both transmission and reflection, thereby enabling integrated functionalities. We establish the first systematic STARS taxonomy, pioneer single-/dual-baseline sensing architectures, and introduce a target-end STARS sensing paradigm. Furthermore, we integrate electromagnetic reconfigurable metasurface design, multi-domain channel modeling, and joint beamforming with edge-coordinated scheduling algorithms. Contribution/Results: The proposed framework achieves a paradigm shift from communication-only to full-stack 6G capabilities, significantly improving spectrum-energy-hardware efficiency: sensing latency is reduced by over 30%, content delivery delay by 40%, and the work actively supports ongoing 3GPP/ITU standardization efforts for STARS.

2 citationsRead paper

Enhancing Control Policy Smoothness by Aligning Actions with Predictions from Preceding States

Jan 26, 2026

This work addresses the challenge of deploying deep reinforcement learning policies in real-world control tasks, where high-frequency action oscillations often degrade performance and stability. Existing approaches typically rely on handcrafted state similarity metrics that fail to accurately capture system dynamics. To overcome this limitation, the paper proposes ASAP (Action Smoothing via Adaptive Proximity), a novel method that constructs “transition-induced similar states” directly from environmental feedback and empirical data. By incorporating action alignment constraints and second-order difference regularization, ASAP effectively suppresses undesirable high-frequency oscillations without requiring heuristic assumptions about state similarity. Experimental results on Gymnasium and Isaac-Lab benchmarks demonstrate that ASAP significantly enhances policy smoothness and overall task performance, outperforming current state-of-the-art methods.

1 citationsRead paper
Recent publications

Latest Papers