我是占俊飞,来自中国,即将于2026年10月加入帝国理工学院计算系攻读博士学位,师从 Giuliano Casale 教授。我近期于宾夕法尼亚大学获得电气工程专业硕士学位,硕士期间的研究由 Saswati Sarkar 教授指导。本科期间的研究分别由 Tengjiao He 教授和 Kwan-Wu Chin 教授指导。
我的研究兴趣主要集中在 optimization、Large Language Models (LLMs) 和数据驱动决策,特别是在能量受限的计算环境中。我主导和参与了多个关于 task scheduling、LLMs 和 green IoT networks 的研究项目。我热衷于将数学建模、机器学习和控制理论相结合,以解决现实世界的系统挑战。
在学术之外,和朋友聚餐、社交是我最喜欢的活动之一。运动也是我生活中重要的一部分,我曾在短跑、攀岩、赛艇和乒乓球方面获得过奖项。
📄 你可以在这里下载我的完整简历。
…计算机科学博士
帝国理工学院,英国
电气工程硕士
宾夕法尼亚大学,美国
应用数学与信息计算科学双学士
伯明翰大学(英国)& 暨南大学(中国)
论文「Trains but Doesn’t Learn: A Post-Training Delivery Benchmark for LLM Agents as Forward-Deployed Engineers」被 The 2026 Conference on Empirical Methods in Natural Language Processing (EMNLP 2026) 录用。
论文「Seeing is Free, Speaking is Not: Uncovering the True Energy Bottleneck in Edge VLM Inference」被 The 34th ACM International Conference on Multimedia (ACM MM 2026) 录用。
论文「Orchestrating Data Collection and Computation in Green IoT Networks」被 IEEE Internet of Things Journal 录用。
合作论文「Graph Learning-based Update Manipulation Attack on Federated Fine-Tuning of LLMs over Wireless Networks」被 The 24th ACM International Conference on Mobile Systems, Applications, and Services (MobiSys '26 Posters) 录用。
获 2026 ESE Department Master's Top 10% GPA Award,表彰在硕士阶段的学术表现。
合作论文「Joint Function Configuration and Multislot Offloading in Solar-Powered Serverless Edge Computing」被 IEEE Internet of Things Journal 录用。
合作论文「SpikeBP: Efficient Spike-Driven Transformer for Blood Pressure Waveform Generation with Frequency Knowledge Distillation」被 IEEE ICASSP 2026 选为 口头报告(Oral)。
Trains but Doesn’t Learn: A Post-Training Delivery Benchmark for LLM Agents as Forward-Deployed Engineers
Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2026. To appear
Seeing is Free, Speaking is Not: Uncovering the True Energy Bottleneck in Edge VLM Inference
Proceedings of the 34th ACM International Conference on Multimedia (ACM MM), 2026. To appear · arXiv:2607.09520
PRISM: Privacy-Aware Routing for Adaptive Cloud–Edge LLM Inference via Semantic Sketch Collaboration
Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), 2026
Orchestrating Data Collection and Computation in Green IoT Networks
IEEE Internet of Things Journal, 2026
Joint Function Configuration and Multi-Slot Offloading in Solar-Powered Serverless Edge Computing
IEEE Internet of Things Journal, 2026
SpikeBP: Efficient Spike-Driven Transformer for Blood Pressure Waveform Generation with Frequency Knowledge Distillation
IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2026 · Oral
Poster: Graph Learning-based Update Manipulation Attack on Federated Fine-Tuning of LLMs over Wireless Networks
Proceedings of the 24th ACM International Conference on Mobile Systems, Applications, and Services (MobiSys Posters), 2026
RL-Enhanced Disturbance-Aware MPC for Fast and Robust UAV Trajectory Tracking
IEEE International Conference on Systems, Man, and Cybernetics (SMC), 2025
Task Offloading and Approximate Computing in Solar Powered IoT Networks
IEEE Networking Letters, 6(1), 26–30, 2024
最新的具体出版物,请关注我的谷歌学术主页。