In recent years, my research has focused on leveraging large language models (LLMs) to enhance the automation and intelligence of AI systems. In the field of reinforcement learning, I address the challenge that traditional reward function design heavily depends on expert knowledge and involves high trial-and-error costs. I explore using LLMs for automated reward modeling, employing code generation and iterative optimization to construct complex reward functions with stronger contextual understanding, thereby improving policy learning efficiency and performance.
Beyond reinforcement learning, I extend my work to computer vision and uncertainty modeling. In visual tasks, I develop diffusion-based interactive image matting methods that enhance fine-grained modeling capabilities through model structure transfer. For safety and robustness, I incorporate Dempster–Shafer evidence theory to detect and defend against adversarial examples through evidence consistency analysis.
1. Principal Investigator, Shanghai University Youth Talent Start-Up Grant (Completed).
1. Third Shanghai University Teaching Innovation Competition (Junior Faculty Group): Excellence Award.
1. Programming and Object-Oriented Conception
2. Machine Learning
3. Data Engineering
4. Theory and Application of Foundation Models
1. Chen, Lunde; Abdellatif, Slim; Berthou, Pascal; Nougnanke, Kokouvi Benoit; Gayraud, Thierry. "A generic and configurable topology discovery service for software defined wireless multi-hop network." Proceedings of the 15th ACM International Symposium on Mobility Management and Wireless Access, 2017: 101-104.
2. Li, Huan; Chen, Lunde. "RSSI-aware energy saving for large file downloading on smartphones." IEEE Embedded Systems Letters 7.2 (2015): 63-66.
3. Chen, Lunde; Abdellatif, Slim; Simo, Armel Francklin; Gayraud, Thierry. "Virtual link embedding in software-defined multi-radio multi-channel multi-hop wireless networks." Proceedings of the 21st ACM International Conference on Modeling, Analysis and Simulation of Wireless and Mobile Systems, 2018: 163-172.
4. Chen, Lunde; Abdellatif, Slim; Gayraud, Thierry; Berthou, Pascal. "A Steiner tree based approach for the efficient support of multipoint communications in a multi-domain context." 2017 IEEE symposium on computers and communications (ISCC), 2017: 316-321.
5. Chen, Lunde; Abdellatif, Slim; Tegueu, Armel Francklin Simo; Gayraud, Thierry. "Embedding and re-embedding of virtual links in software-defined multi-radio multi-channel multi-hop wireless networks." Computer Communications 145 (2019): 161-175.
6. Chen, Lunde; Abdellatif, Slim; Chakroun, Raoua. "Point-to-multipoint Virtual Link Embedding in Multi-domain SDN Networks." 2019 IEEE International Conference on Advanced Networks and Telecommunications Systems (ANTS), 2019: 1-6.
7. Dong, Lele; Chen, Lunde; Kong, Suyao; Gu, Siyuan; Liu, Wanyu; Du, Shuimiao. "Set-valued medical image classification with evidential CNN: a first test with Covid-19 dataset." 2022 16th IEEE International Conference on Signal Processing (ICSP) 1 (2022): 463-467.
8. Tan, Xiaodong; Du, Jinxin; Chen, Lunde; Liu, Wanyu. "A Novel Deep Q-Network-Based Scheme for Online Virtual Link Embedding in Software Defined Networks." 2023 8th International Conference on Cloud Computing and Big Data Analytics (ICCCBDA), 2023: 516-522.
9. Wu, Licheng; Xu, Yue; Dai, Zhicheng; Wang, Rui; Chen, Lunde. "Intrusion detection with set-valued classification: Leveraging dempster-shafer theory and deep learning." 2023 9th International Conference on Computer and Communications (ICCC), 2023: 2155-2161.
10. Gu, Siyuan; Dai, Zhicheng; Wu, Licheng; Liu, Wanyu; Chen, Lunde. "Dempster-Shafer theory empowered deep learning for enhanced 3D PET-CT medical image segmentation: a first test on automated lesion segmentation in whole-body dataset." 2024 4th International Conference on Neural Networks, Information and Communication Engineering (NNICE), 2024: 350-354.
11. Gan, Yucheng; Dai, Zhicheng; Wu, Licheng; Liu, Wanyu; Chen, Lunde. "Deep Reinforcement Learning and Dempster-Shafer Theory: A Unified Approach to Imbalanced Classification." 2023 3rd International Conference on Computer Science, Electronic Information Engineering and Intelligent Control Technology (CEI), 2023: 67-72.
12. Zhu, Xinning; Du, Jinxin; Fu, Qiongying; Chen, Lunde. "LLM-Based Reward Engineering for Reinforcement Learning: A Chain of Thought Approach." 2025 10th International Conference on Cloud Computing and Big Data Analytics (ICCCBDA), 2025: 222-227.
13. Huang, Longfei; Liang, Yu; Zhang, Hao; Chen, Jinwei; Dong, Wei; Chen, Lunde; Liu, Wanyu; Li, Bo; Jiang, Peng-Tao. "Sdmatte: Grafting diffusion models for interactive matting." Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2025: 15229-15239.
14. Xu, Yue; Qiu, Yibo; Huang, Ling; Wang, Rui; Chen, Lunde; Zhou, Jiwen; Liu, Wanyu. "Adversarial example detection and defense based on the evidence consistency from Dempster-Shafer layers." Knowledge-Based Systems (2025): 114937.
15. Huang, Longfei; Liang, Yu; Zhang, Hao; Zhu, Xinning; Chen, Lunde. "Diffusion for automatic matting." Neurocomputing (2026): 133344.
