Daven's Homepage

Cheng Deng @ Daven

Ph. D. Candidate & LLMer

Acemap IIoT, SEIEE
Shanghai Jiao Tong University

Email: davendw@sjtu.edu.cn
Office: SEIEE 1-432, SJTU
WeChat: davendw

Blog

GitHub

Google Scholar

About me

Currently, I'm pursuing my doctoral degree in Computer Science in Acemap, IIOT Lab, at Shanghai Jiao Tong University, where I have the honor of being supervised by Prof. Weinan Zhang, Prof. Luoyi Fu and Prof. Xinbing Wang. I was also selected into Wenjun Wu Honored Ph.D. Class in 2021.

My research interests include natural language processing, knowledge graphs, and data mining. Specifically, I am devoted to the research of specialized domain large language models (collaborating with Prof. Junxian He and Prof. Zhouhan Lin) and utilizing AI techniques to improve geoscience and design practices as reliable AI systems (collaborating with Prof. Ying Wen).

What's more, I am in charge of several IIOT primary services:

For research:

For laboratory:

What's News

[2023-09] Coming soon!! PDF Parser toolkits for text corpus pre-processing!

[2023-09] Coming soon!! GeoGalactica: A Larger language model for geoscience!

[2023-08] New!! Invited talk sharing in UGA about geoscience foundation model k2 and GeoAI.

[2023-06] New!! We propose a large language model for geoscience knowledge understanding and utilization, the paper is under review while the model and related data are open-sourced on davendw49/k2!

[2022-12] Cooperation with CAS Team, we analysis the 'Paperdemic' phenomenon during the COVID-19 pandemic and the paper has been accepted by European Journal of Internal Medicine (SCI-Q1)!

[2022-11] Our GeoScience Academic Knowledge Graph and its related systems and resources GAKG for DDE is promoted at IUGS in Paris!

[2022-11] Navigation system over GAKG is newly developed and publicly accessible GeoScience Knowledge Navigation System is publicly accessible! [Bilibili, Youtube]

[2022-05] Our COVID-19 Interdisciplinary Research System Covidia is publicly accessible!

[2022-01] Our Semantic search system is coming, a re-organized system of GAKG.

[2021-08] Our GeoScience Knowledge Platform GAKG Platform is publicly accessible!

[2021-08] Our paper "GAKG: A Multimodal GeoScience Academic Knowledge Graph" has been accepted as a resource paper at CIKM-2021!

Publication [Google Citation]

[ArXiv] Cheng Deng, Tianhang Zhang, Zhongmou He, Yi Xu, Qiyuan Chen, Yuanyuan Shi, Luoyi Fu, Weinan Zhang, Xinbing Wang, Chenghu Zhou, Zhouhan Lin and Junxian He, "Learning A Foundation Language Model for Geoscience Knowledge Understanding and Utilization" [Under Review]. [Paper] [Code] [Demo]

[ArXiv] Cheng Deng, Jiaxing Ding ,Weinan Zhang, Luoyi Fu, Xinbing Wang, Chenghu Zhou. "Covidia: COVID-19 Interdisciplinary Academic Knowledge Graph Construction and Application" [Under Review].

[CCF-B] Cheng Deng, Fan Xu, Jiaxing Ding, Weinan Zhang, Luoyi Fu, Xinbing Wang, Chenghu Zhou, et al. "FMGNN: Fused Manifold Graph Embedding via Geometry Coreset" in TKDD, 2023 [Under Review]. [Paper]

[HSSCOMMS-2023] Yuanbo Luo, Lizhi Tao, Fan Xiao, Cheng Deng, Xinbing Wang, Chenghu Zhou, et al. "The impacts of COVID-19 on gender inequality in academia" in HSSCOMMS. [Minor Revision].

[SCI-Q1] Yang Yang, Na Zhao, Ting Ma, Ze Yuan, Cheng Deng, et al. "'Paperdemic' during the COVID-19 pandemic" in "Letter to the Editor" of European Journal of Internal Medicine, 2022, DOI: 10.1016/j.ejim.2022.09.030.

[CCF-B] Cheng Deng, Yuting Jia, Weinan Zhang, Luoyi Fu, Xinbing Wang, Chenghu Zhou, et al. "GAKG: A Multimodal GeoScience Academic Knowledge Graph" in CIKM, 2021, DOI: 10.1145/3459637.3482003. [Website] [Paper] [Code] [Slide]

[Arxiv] Cheng Deng, Bo Tong, Luoyi Fu, Jiaxin Ding, Dexing Cao, Xinbing Wang and Chenghu Zhou. "PK-Chat: Pointer Network Guided Knowledge Driven Generative Dialogue Model". [Paper] [Data] [Code]

[CCF-B, SCI-Q2] Jiaqi Liu, Zhiwen Yu, Bin Guo, Cheng Deng, Luoyi Fu, Xinbing Wang, Chenghu Zhou. "EvolveKG: A General Framework to Learn Evolving Knowledge Graphs" in Frontiers of Computer Science, 2022. [Paper] [Code]

[CCF-A] Jiaqi Liu, Cheng Deng, Luoyi Fu, Xinbing Wang, et al. "Evolving Bipartite Model Reveals the Bounded Weights in Mobile Social Networks: A Case Study in Recommendation Network" in IEEE Transactions on Mobile Computing, 2020, DOI: 10.1109/TMC.2020.3017630. [Paper] [Link] [Code]

Projects

K2: The first geoscience Large Language Model, sharing by Prof. Cleverley [Github].

Navigation over GeoScience Academic Knowledge Graph, an introduction video is shown below, audio by Cinna (Qiyuan Chen from University of Waterloo) [Website]

[GeoScience Multimodal Knowledge Graph] 图文数一体化的地学知识图谱构建与应用, 专项项目, Construction and application of geoscience knowledge map with integrated graphics, text and data. [Website]

[Knowledge Graph and Graph Analysis] 知识驱动下的大图度量与分析, 国际(地区)合作与交流项目, Knowledge-driven big picture measurement and analysis.

[Computer science and COVID-19] 新型冠状肺炎病毒扩散溯源和疫情防预关键技术研究, 专项项目, 面上项目, Research on key technologies of novel coronavirus spread traceability and epidemic prevention.
[Interdisciplinary Research during the COVID-19] Knowledge Graph for COVID-19 related Interdisciplinary Research. [Website].

[Knowledge Driven Design Material Recommendation System] Construction a design common sense knowledge graph and deploy the recommendation system via reasoning over knowledge graph. Cooperated with HUAWEI.

[Ancient Chinese Literature] I put forward an ancient Chinese character editor that automatically summarized the ancient Chinese characters in the plate that had been extracted out. I am currently working on the second stage of development and named it "shuowen".

DDE Deep-time Digital Earth[Project Page], contribute to the projects as follows

Teaching [TA] and Services

TA

[Mobile Networks] [2019,2020,2021,2022]

[IEEI (B)] [2020] [2021,2022,2023]

[Program Design (C++)] [2020]

[Data Science CS245-2] Kaggle

Reviewer

Reviewer for Journal TKDE [2019-]

Reviewer for Journal TMC [2020]

Reviewer for Journal FCS [2023]

Talks for Academic Sharing

Invited by Prof. Gengchen Mai from UGA, I gave a talk with on topic of "large language model in geoscience" in the seminar of GeoAI, Aug. 2023.

Attending DDE Conference hosted by China University of Geosciences (Beijing), I gave a talk on "Building a sustainable text corpus for geoscience", Sept. 2023.

Awards

Being selected into Mentor Group of Guozhi Class

Wu Wen Jun Honorary Doctoral Scholarship, 2020.

Honored Ph.D Wenjun Wu Honored Ph.D Class, 2020.

Outstanding Undergraduate Thesis in Hunan Province, 2019 (Top 1%).

Outstanding Undergraduate Thesis in Hunan University, 2019 (Top 1%).

Meritorious Winner in Mathematical Contest In Modeling (MCM), 2018 (Top 7.09%).

Ten Influential Students of Hunan University Student Association Federation, 2018.

3 × First Class Scholarship for Undergraduate Student in Hunan University, 2016-2019 (Top 5%).