Duan, Chenrui (段晨瑞)
 |
PhD Student, |
About me
I am a Ph.D. candidate in the College of Engineering at Westlake University and Zhejiang University (ZJU), under the supervision of Prof. Stan Z. Li (IEEE Fellow). I received my B.S. degree from Hebei University of Technology (HeBUT) in 2020. My research focuses on the development and application of advanced artificial intelligence techniques, particularly large language models, multimodal models, and autoregressive generative approaches, with an emphasis on leveraging these methods for innovative solutions in biotechnology.If you would like to discuss potential collaborations or learn more about my work, please feel free to contact me.
Research Interests
AI for Science
Large Language Model Applications
Multimodal Models
Autoregressive Generative Models
Selected Publications
. PhyloGen: Language Model-Enhanced Phylogenetic Inference via Graph Structure Generation. Accepted by NeurIPS 2024.
Proposed a generative phylogenetic inference model, utilizing a pre-trained language model and graph neural network to generate tree structures with linear complexity in branch length optimization. The model uses a scoring function to guide gradient optimization, accelerating convergence and improving computational efficiency.
FGBERT: Function-Driven Pre-trained Gene Language Model for Metagenomics.
Developed a large pre-trained language model based on 100 million metagenomic sequences, incorporating gene masking and triplet contrastive learning strategies to improve understanding of unknown gene regions. Built the first benchmark tool for this field, addressing complex functional relationships among sequences.
A Review of Artificial Intelligence based Biological-Tree Construction: Priorities, Methods, Applications and Trends.
Reviewed the application of generative models and large language models in phylogenetic inference, explored multimodal data fusion strategies for biological tree construction, and proposed directions for optimizing multimodal feature representation and generation capabilities.
Must: Maximizing Latent Capacity of Spatial Transcriptomics Data. Accepted by BIBM (CCF-B), 2024.
Developed a multimodal fusion model based on GNN and manifold learning, integrating gene expression and spatial data to construct a unified latent representation, improving analysis performance by 10%.
GVFOM: a novel external force for active contour based image segmentation[J]. Information Sciences, 2020, 506: 1-18. (SCI Region I, IF 5.524, supervisor's first work)
Image Segmentation Using Active Contours with Hessian Based Gradient Vector Flow External Force[J]. Pattern Analysis and Appications, 2020. (SCI Zone 3, IF 3.245, supervisor's first work)
Video Captioning in Compressed Video. International Conference on Pattern International Conference on Pattern Recognition, 2020.
Academic service
Reviewer
IEEE Transactions on Neural Networks and Learning Systems
IEEE Transactions on Industrial Informatics
ACM Transactions on Knowledge Discovery from Data
IEEE Access
Selected Honors and Awards
In September 2019, I obtained the qualification of “excellent camper” in the College of Engineering of Westlake University, and at the same time, I carried out the research related to the frontier topics of artificial intelligence one year ahead of schedule.
During this period, I have completed one thesis and two invention patents with the organization of Westlake University.
In September 2020, I was awarded the freshman scholarship of Westlake University Engineering School, only 2 students in cs direction.
I have been actively assisting the laboratory to complete the cooperative research with governmental departments and cooperative enterprises.
In October 2020, I joined the student team of Artificial Intelligence and Robotics Laboratory of Xihu University, and represented Xihu University to compete with the teams from University of Technology Sydney, Alibaba, Zhejiang University, Renmin University of China, and other famous domestic and foreign colleges and universities and large companies in the Zhijiang Cup Global Artificial Intelligence Algorithm Competition.
Our team ranked 2nd among 214 teams and won the second prize, which expanded the influence and popularity of Xihu University in the field of artificial intelligence to a certain extent.
In the 2020 academic year, I participated in the “Beauty of Life” joint creation camp of Xihu University and China Academy of Art, in which the neural network simulated the thinking process of the human brain, and the process and the final result were shown through artistic creation at the same time, so that we could see how the network analyzes and understands the pictures and videos in the specific tasks of image and video analysis. I am a member of the Zhejiang University of Science and Technology.
In 2020, I joined the Graduate Art Troupe of Zhejiang University, and participated in the New Year's Day Party of Zhejiang University and the New Year's Day Party of the School of Computer Science.