喜讯 | 香港中文大学（深圳）首届研究生科研论坛文理科口头学术报告金奖得主出炉
榜样领航，逐光而行
香港中文大学（深圳）研究生院举办的首届研究生科研论坛已圆满落下帷幕，各位研究生们的学术之路仍在继续。在此，让我们回顾科研论坛中优秀获奖者的闪耀成果，希望鼓励在校硕博生和未来的研究生们，怀着对知识的渴望、对创新的追求，砥砺前行，探索更多未知科研领域。在这片充满智慧和创新的土壤上，努力耕耘，书写属于自己的学术辉煌篇章！
在科研论坛活动当天，刘岳臻和晏志远获得理工学科组口头学术报告金奖，姜帅宇和曹哲获得人文社会学科组口头学术报告金奖。另外，三名同学获得了文科组银奖，七名同学获得了理科组银奖。论坛另设有优秀海报奖五名，最具人气海报奖一名。祝贺各位获奖的研究生！
理工学科组 口头学术报告金奖
刘岳臻（2021级博士研究生）
“我现在是香港中文大学（深圳）理工学院计算机与信息工程专业三年级博士生，导师是俞江帆教授。我的研究兴趣包括基于视觉的智能控制和微型机器人的生物医学应用。”
演讲课题：
Automatic Navigation of Microswarms for Dynamic Obstacle Avoidance
课题摘要：
Control and navigation of microrobotic swarms have drawn extensive attention recently. Avoiding dynamic obstacles using swarms is one of the major challenges that still remain unsolved. In this work, we develop a control strategy to navigate microrobotic swarms to targeted positions while avoiding dynamic obstacles. We first propose a criterion to evaluate the real-time locomotion efficiency during dynamic obstacle avoidance, i.e., the swarm moving direction and the distance between the swarm and the target. Subsequently, a hierarchical radar with three functional boundaries (detection, safety, and prediction circle) is designed for swarms. The optimal moving direction of the swarm with the existence of dynamic obstacles is selected based on the three circles. The effectiveness of the algorithm is validated by simulations and experiments. Using the proposed strategy, the swarm is capable of avoiding multiple moving obstacles and reaching the predefined target. Finally, to show the compatibility of the proposed control method, the swarm is deployed in a micromaze with different dynamic obstacles, and the results also validate the effectiveness of the strategy.
理工学科组 口头学术报告金奖
晏志远（2022级授课型硕士研究生）
“我来自香港中文大学（深圳）数据科学学院数据科学专业，师从数据科学学院吴保元教授。目前已在NeurIPS、ICCV、Bioinformatics等国际会议/期刊上发表论文。主要研究方向为AIGC Detection、Face Forgery Detection、AI4Science。此前曾在大疆、百度、深圳市大数据研究院、腾讯AILab进行科研工作和学习。”
演讲课题：
A Generalizable Approach to Face Forgery Detection and the Introduction of DeepfakeBench
课题摘要：
This talk centers on the identification of manipulated images, with a special emphasis on face image forgeries, commonly known as deepfake. We address the challenge of generalization in this arena and spotlight research centered on a disentanglement-based approach, which leverages exclusively forgery-shared features for more general deepfake detection. Additionally, we will present the significant contributions of our innovative DeepfakeBench, a robust benchmarking tool designed for deepfake detection.
人文社会学科组 口头学术报告金奖
姜帅宇（2022级授课型硕士研究生）
“我是香港中文大学（深圳）经管学院金融学专业的研究生，导师是陈齐辉教授和张劲帆教授”
演讲课题：
Asset Pricing in China’s Stock Market
课题摘要：
This paper uses traditional machine learning methods and deep neural networks based on both firm-specific characteristics and macroeconomic variables to price China’s A-share stock market. The main contributions are as follows: We give the stochastic discount factor a flexible form and compare different models’ performances. Since the Chinese government adopts various policies to maintain financial stability, we borrow the idea from a generative adversarial network to find the true SDF by selecting a moment condition that minimizes return volatility. Additionally, we compare this model’s performance with Chen’s work and find that this model can obtain a higher Sharpe ratio and R2.
人文社会学科组 口头学术报告金奖
曹哲（2022级授课型硕士研究生）
“我来自香港中文大学（深圳）人文社科学院中国语言文学专业研究生一年级，导师是邢文教授。”
演讲课题：
Where Is the Gold on the High Terrace: The Naming and Textualization of a Place in Cultural History/高台何处有黄金——“黄金台” 典故的生成与流变探析
课题摘要：
本文研究了“黄金台”，作为承载了一段史事的古代建筑，在文学史中被重新命名、文本化和典故化的过程，从而被建构成古典诗歌中一个典型的“跨文本”景观，并由此引申，探讨了文与史、名与实的关系等一系列问题。
恭喜以下研究生们获得“口头学术报告银奖”！
此外，评委老师对学术海报展进行评分，最终有五位同学获得“优秀学术海报奖”。在为期一周的学术海报展投票的过程中，共收到我校同学和教职员工186人次有效投票。其中，来自理工学院计算机与信息工程哲学硕士项目的胡凯源以21%的投票率获得首届研究生科研论坛“最具人气海报奖”。一起来看看部分同学的研究成果吧！
余湉 优秀海报奖获得者
2022级 医学院 生物科学专业博士生
导师：刘国珍教授
课题：CRISPR/Cas12a-based immunosensors on magnetic beads for rapid cytokine detection: IL-6 as an example
摘要：The rapid detection of biomarkers like nucleic acids and proteins in clinical samples can significantly enhance clinical outcomes in chronic diseases through early diagnosis and prevention. The Clustered Regularly Interspaced Short Palindromic Repeat (CRISPR) technology, coupled with the Cas protein, represents an innovative biosensing approach, renowned for its remarkable accuracy, specificity, and sensitivity. This technology has effectively harnessed the success in the detection of a broad range of analytes such as nucleic acids, proteins, et al. Cytokines are signally molecules between immune cells and biomarkers for many diseases. It is challenging to detect cytokines. In this study, taking advantage of magnetic beads and immunoassays, we developed a CRISPR/Cas12a-based immunosensors on magnetic beads, which exhibits rapid and highly sensitive detection of interleukin-6 (IL-6) down to 0.1 pg/ml within a compact reaction system excluding the need for additional sample purification or amplification steps. This breakthrough underscores its potential as a formidable alternative for future protein detection in both industrial and medical applications.
祁卫敏 优秀海报奖获得者
2019级 理工学院 计算机与信息工程博士生
导师：钱辉环教授
课题：Environmentally Adaptable Station Keeping of Sailing Robots with Application to Ocean Observation
摘要：Under the background of the following parts: (1) Exploring the vast ocean environment has a wide range of application prospects. (2) Sailing robots have several promising features (e.g., long-range sailing, energy-saving and low-noise) to perform the long-term observation task. (3) Aero and hydrodynamic disturbances from environment make this task extremely challenging. We aim to realize two objectives: (1) In a given area the sailing robot will sail smoothly and collect valid data through continuous reciprocation. Both control stability and sailing safety should be guaranteed. (2) The robot will leverage environmental disturbances (e.g., wind and wave) as an energy reservoir for observation of a wider ocean area.
胡凯源 最具人气海报奖获得者
2022级 理工学院
计算机与信息工程研究型硕士生
导师：王方鑫助理教授
课题：Understanding User Behavior in Volumetric Video Watching: Dataset, Analysis and Prediction
摘要：Volumetric video has emerged as a new attractive video paradigm in recent years since it provides an immersive and interactive 3D viewing experience with six degrees of freedom (DoF). Unlike traditional 2D or panoramic videos, volumetric videos require dense point clouds, voxels, meshes, or huge neural models to depict volumetric scenes, which results in a prohibitively high bandwidth burden for video delivery. Users' behavior analysis, especially the viewport and gaze analysis, then plays a significant role in prioritizing the content streaming within users' viewport and degrading the remaining content to maximize user QoE with limited bandwidth. Although understanding user behavior is crucial, to the best of our best knowledge, there are no available 3D volumetric video viewing datasets containing fine-grained user interactivity features, not to mention further analysis and behavior prediction. In this paper, we for the first time release a volumetric video viewing behavior dataset, with a large scale, multiple dimensions, and diverse conditions. We conduct an in-depth analysis to understand user behaviors when viewing volumetric videos. Interesting findings on user viewport, gaze, and motion preference related to different videos and users are revealed. We finally design a transformer-based viewport prediction model that fuses the features of both gaze and motion, which is able to achieve high accuracy at various conditions. Our prediction model is expected to further benefit volumetric video streaming optimization.
结 语
最后，再次祝贺香港中文大学（深圳）首届研究生科研论坛的所有获奖者！
同学们的研究成果让我们了解到科研思维的碰撞与学术智慧的融合，呈现了一场充满激情和创新的学术盛会。
希望本次研究生科研论坛能激励同学们的科研热情，提高务实创新的学术能力，积极与来自不同研究背景和领域的伙伴，一起探讨科研知识，激发碰撞出新的学术思维火花，促进学术的多元化跨领域的深度交流。
期待我们2024年研究生科研论坛再相会！
获取更多资讯
请关注香港中文大学（深圳）研究生院公众号