Congratulations to Master's Students Guangyin Bao and Zixuan Gong, Supervised by Professor Duoqian Miao, on Receiving the 2025 Outstanding Master's Thesis Award from Tongji University
Under the supervision of Professor Duoqian Miao, two master students, Guangyin Bao and Zixuan Gong, have been awarded the 2025 Outstanding Master's Thesis Award at Tongji University. Both theses focus on brain visual decoding using functional magnetic resonance imaging (fMRI), a cutting-edge interdisciplinary field bridging neuroscience and artificial intelligence. Bao’s research explores multi-subject collaborative encoding and decoding models with privacy preservation, while Gong’s work advances multimodal decoding tasks including image retrieval and video reconstruction. Their innovative contributions have been published in top-tier conferences such as NeurIPS, ICLR, AAAI, and ACM MM. These achievements highlight the team's leading research capability in brain-inspired computing and neural decoding.
祝贺苗夺谦教授指导的硕士生包广垠、龚子璇荣获同济大学2025年优秀硕士学位论文称号。两位同学的研究均聚焦于功能磁共振成像（fMRI）大脑视觉解码这一神经科学与人工智能交叉领域的前沿方向，从不同技术路径推动了脑启发人工智能研究的发展，充分体现了团队在该方向上的深厚积累与持续突破。
包广垠同学的论文《多被试协同的多模态fMRI编解码算法研究》围绕fMRI视觉编解码任务，针对数据噪声高、被试差异大及隐私安全风险等关键问题，提出了多被试协同建模、联邦学习隐私保护机制以及基于生成模型的高保真编码方法，有效提升了模型的精度、泛化能力和安全性。相关成果发表于人工智能CCF-A类顶级会议AAAI 2025和ICLR 2025。龚子璇同学的论文《针对fMRI的多模态大脑解码算法研究》则聚焦于图像检索、图像重建与视频重建等多模态解码任务，创新性地引入频域表征学习、低秩适配器跨被试建模以及基于扩散模型的视频生成框架，在多个国际基准任务上取得领先性能，相关成果发表于NeurIPS 2024、ACM Multimedia 2024及AAAI 2025等CCF-A类顶级会议。两位同学的研究从不同切入点共同深化了对人脑视觉信息处理机制的理解，并为下一代脑机接口与智能计算系统提供了重要技术支撑。
此次两位同学同时荣获校级优秀硕士学位论文，既是对其个人科研能力与创新成果的高度肯定，也彰显了团队在脑科学与人工智能交叉研究方向上的科研实力与育人成效。未来，团队将继续围绕脑启发智能与神经计算开展深入探索，力争在该前沿领域取得更多突破性进展。
