BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Department of Electrical and Computer Engineering (HKUECE) 電機與計算機工程系 - ECPv6.16.3//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:Department of Electrical and Computer Engineering (HKUECE) 電機與計算機工程系
X-ORIGINAL-URL:https://ece.hku.hk
X-WR-CALDESC:Events for Department of Electrical and Computer Engineering (HKUECE) 電機與計算機工程系
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:Asia/Hong_Kong
BEGIN:STANDARD
TZOFFSETFROM:+0800
TZOFFSETTO:+0800
TZNAME:HKT
DTSTART:20250101T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20260514T133000
DTEND;TZID=Asia/Hong_Kong:20260514T143000
DTSTAMP:20260611T045558
CREATED:20260505T021413Z
LAST-MODIFIED:20260505T021413Z
UID:115830-1778765400-1778769000@ece.hku.hk
SUMMARY:RPG Seminar – Cooperative Edge AI: From Event-triggered Inference to Efficient Model Downloading
DESCRIPTION:Zoom Link \nhttp://hku.zoom.us/j/7074144117?omn=95813783034 \nAbstract\nCooperative edge AI enables edge devices and edge servers to collaboratively execute intelligent tasks under limited computation\, storage\, energy\, and communication resources. In this talk\, we discuss two complementary research directions toward communication-efficient cooperative edge AI. First\, we introduce an event-triggered cooperative inference framework for rare-event detection in edge intelligence systems. Rare events are usually infrequent but highly critical\, while conventional edge inference systems may overlook them due to data imbalance and rigid resource allocation. To address this issue\, a dual-threshold multi-exit architecture is adopted\, allowing confident normal events to be processed locally while complex or uncertain rare events are selectively offloaded to the edge server for more accurate classification. Second\, we present an efficient AI model downloading framework based on parametric-sensitivity-aware retransmission. Instead of treating all model parameters equally\, this framework exploits the unequal importance of neural network parameters and allocates wireless retransmission resources to more sensitive model packets. In this way\, downloading latency can be reduced while inference performance is preserved. The talk concludes with a discussion of future research directions in cooperative edge AI\, highlighting open challenges and opportunities in communication-efficient inference\, adaptive model deployment\, and resource-aware edge intelligence. \nSpeaker\nMr Zhou You\nDepartment of Electrical and Computer Engineering\nThe University of Hong Kong \nBiography of the Speaker\nZhou You is currently pursuing a Ph.D. degree in the Department of Electrical and Electronic Engineering at The University of Hong Kong\, under the supervision of Prof. Kaibin Huang. He received his B.Eng. degree in Electrical Engineering from the University of Wisconsin–Madison\, USA\, in 2021. His research interests include wireless communications\, edge inference\, and AI model downloading. \nOrganiser\nProf. Kaibin HUANG \nDepartment of Electrical and Computer Engineering\, The University of Hong Kong \nAll are welcome.
URL:https://ece.hku.hk/events/20260514/
CATEGORIES:Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2024/11/rpg-seminar.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20260515T140000
DTEND;TZID=Asia/Hong_Kong:20260515T150000
DTSTAMP:20260611T045558
CREATED:20260512T081414Z
LAST-MODIFIED:20260512T081414Z
UID:115938-1778853600-1778857200@ece.hku.hk
SUMMARY:RPG Seminar – Stabilizing Streaming Video Geometry via Dynamic Feature Normalization
DESCRIPTION:Zoom Link \nhttps://hku.zoom.us/j/94833409754?pwd=ER6VaveQbdEzOQzhuFKThpSdRusUDs.1 \nAbstract\nConsistent 3D geometry estimation from streaming RGB input is crucial for real-world applications such as autonomous driving\, embodied AI\, and large-scale reconstruction. \nWhile modern monocular geometry foundation models achieve strong single-image accuracy\, they exhibit severe temporal inconsistency on continuous input\, notably dominated by scale–shift drifting. Through targeted empirical analysis\, we trace this instability to its root cause: fluctuations in latent feature statistics\, whose mean and variance directly determine the predicted depth’s scale and shift. Building on this insight\, we introduce Dynamic Feature Normalization (DyFN)\, a lightweight\, causal recurrent module that dynamically and robustly modulates feature statistics to maintain stable geometry over time. We adapt powerful pretrained monocular geometry models for streaming by finetuning only DyFN\, a mere 2% additional parameters\, while keeping the backbone frozen\, thereby achieving temporal consistency without compromising single-image accuracy. Extensive experiments across four benchmarks show that DyFN effectively eliminates temporal artifacts such as disjointed layering and positional jitter\, and achieves state-of-the-art temporal stability\, improving over prior streaming methods by up to 14% and even outperforming heavier non-causal video baselines. \nSpeaker\nMr Xiaoyang LYU\nDepartment of Electrical and Computer Engineering\nThe University of Hong Kong \nBiography of the Speaker\nXiaoyang Lyu is a fourth-year PhD student in the CVMI Lab at the University of Hong Kong\, where he is supervised by Prof. Xiaojuan Qi. He holds a Master’s degree from Zhejiang University and a Bachelor’s degree from the Harbin Institute of Technology.\nXiaoyang’s research focuses on bridging the gap between physical and digital environments by replicating complex physics\, geometry\, and material properties within simulators. He is driven by the conviction that high-fidelity world modeling is essential for advancing embodied AI and developing agents that can effectively assist in the real world. \nOrganiser\nProf Xiaojuan QI \nDepartment of Electrical and Computer Engineering\, The University of Hong Kong \nAll are welcome.
URL:https://ece.hku.hk/events/20260515-2/
CATEGORIES:Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2024/11/rpg-seminar.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20260515T160000
DTEND;TZID=Asia/Hong_Kong:20260515T170000
DTSTAMP:20260611T045558
CREATED:20260511T020048Z
LAST-MODIFIED:20260511T020048Z
UID:115888-1778860800-1778864400@ece.hku.hk
SUMMARY:RPG Seminar – Unveiling the Relationship Between Cation Content and Zeta Potential of Colloids for Forming High-Quality Perovskites
DESCRIPTION:Zoom Link \nhttps://hku.zoom.us/j/94017530661 \nAbstract\nThere are numerous studies focusing on the crystallization dynamics of perovskite materials. However\, the change of precursor properties which can also significantly affect crystallization behavior\, is always ignored. In this seminar\, we establish a comprehensive understanding of the relationship between A-site cations content and zeta potential of precursor\, revealing its influence on perovskite formation and crystallization dynamics. Through in-situ photoluminescence (PL) and X-ray diffraction (XRD) analyses\, we demonstrate how zeta potential impacts the formation process and crystallization behavior of perovskites. Furthermore\, we explore the effects of zeta potential on the optical and electrical properties of the resulting materials. Our findings indicate that achieving a zeta potential near zero facilitates the fabrication of high-quality and additive-free perovskites\, leading to enhanced performance in perovskite solar cells (PSCs) and perovskite light-emitting diodes (PeLEDs). This work provides vital insights into tuning interfacial properties for improved perovskite optoelectronic devices. \nSpeaker\nMr. Qi XIONG\nDepartment of Electrical and Computer Engineering\nThe University of Hong Kong \nBiography of the Speaker\nQi Xiong received the B.S. degree in Polymer Materials and Engineering from Hainan University\, and the M.S. degree in Material Science and Engineering from South China University of Technology. He is currently pursuing the Ph.D. degree in the Department of Electrical and Computer Engineering\, Faculty of Engineering\, The University of Hong Kong. His current research interests include perovskite synthesis and blue perovskite light-emitting diodes (PeLEDs). \nOrganiser\nProf. Wallace C.H. CHOY \nDepartment of Electrical and Computer Engineering\, The University of Hong Kong \nAll are welcome.
URL:https://ece.hku.hk/events/20260515/
CATEGORIES:Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2024/11/rpg-seminar.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20260518T160000
DTEND;TZID=Asia/Hong_Kong:20260518T170000
DTSTAMP:20260611T045558
CREATED:20260513T040826Z
LAST-MODIFIED:20260513T062829Z
UID:115975-1779120000-1779123600@ece.hku.hk
SUMMARY:RPG Seminar – Robust Multivariate Autoregressive Model Estimation under Impulsive Noise
DESCRIPTION:Zoom Link \nhttps://hku.zoom.us/j/9995553636?omn=96185323037 \nAbstract\nThis seminar presents robust multivariate autoregressive (MVAR) model estimation under impulsive noise\, with a focus on extending bias-compensated instrumental-variable methods to contaminated multichannel observations. MVAR models are widely used in time-series prediction\, system identification\, biomedical signal analysis\, and sensor-array processing. However\, conventional estimators can suffer from systematic bias under correlated measurement noise\, and their performance can degrade severely when rare but large-amplitude impulsive samples dominate covariance and correlation statistics. \nThe talk first reviews the transition from AR to MVAR modeling and explains how measurement noise enters the estimation problem. It then introduces the extended instrumental-variable bias-compensation (EIV-BC) framework for correlated noise and discusses why impulsive contamination presents an additional challenge. A robust batch EIV-BC strategy is presented\, using bidirectional estimation and residual-based outlier exclusion to identify unreliable samples before recomputing model statistics. \nExperiments are conducted on both simulated MVAR signals and real-world UCI air-quality sensor data. The results show that impulsive noise significantly degrades standard EIV-BC prediction\, while the robust extension provides more stable estimates and lower prediction errors on non-impulsive test samples. The seminar demonstrates how robust sample selection can improve MVAR estimation reliability in practical noisy sensing environments. \nSpeaker\nMr Mingxi LYU\nDepartment of Electrical and Computer Engineering\nThe University of Hong Kong \nBiography of the Speaker\nMingxi Lyu received the B.S. degree in Mechanical Engineering from Xi’an Jiao tong University in 2019\, and the M.S. degree in Mechanical Engineering from Xi’an Jiao tong University in 2022. He is currently pursuing the Ph.D. degree in electrical and electronic engineering at the Department of Electrical and Electronic Engineering\, The University of Hong Kong. His current research interests include multivariate autoregressive regression and statistical robust estimation of chirp signal with their applications. \nOrganiser\nProf. Shing Chow CHAN \nDepartment of Electrical and Computer Engineering\, The University of Hong Kong \nAll are welcome.
URL:https://ece.hku.hk/events/20260518/
CATEGORIES:Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2024/11/rpg-seminar.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20260519T110000
DTEND;TZID=Asia/Hong_Kong:20260519T120000
DTSTAMP:20260611T045558
CREATED:20260518T030643Z
LAST-MODIFIED:20260518T031123Z
UID:116136-1779188400-1779192000@ece.hku.hk
SUMMARY:RPG Seminar – Toward World Models that are Consistent\, Physically-Grounded\, and Causal
DESCRIPTION:Zoom Link \nhttps://hku.zoom.us/j/98470480541?pwd=loAm99UUdaZEqJZNVRfb9d60al7TZe.1 \nAbstract\nWorld models that simulate future states from past observations are increasingly central to video generation and embodied AI. Yet most existing approaches fall short along two fundamental axes: maintaining temporal consistency over long rollouts\, and grounding spatial reasoning in 3D geometry. \nThis talk presents two research projects addressing each of these challenges\, working toward world models that are both temporally consistent and physically grounded — two properties we argue are essential prerequisites for tackling causality and action understanding. \nSpeaker\nMiss Xiaoshan WU\nDepartment of Electrical and Computer Engineering\nThe University of Hong Kong \nBiography of the Speaker\nXiaoshan Wu is a third-year PhD student in the CVMI Lab at the University of Hong Kong\, where she is supervised by Prof. Xiaojuan Qi. \nXiaoshan Wu’s research focuses on long-horizon\, 3D-aware world models\, with an emphasis on temporal consistency and physical grounding. \nOrganiser\nProf Xiaojuan QI \nDepartment of Electrical and Computer Engineering\, The University of Hong Kong \nAll are welcome.
URL:https://ece.hku.hk/events/20260519-1/
CATEGORIES:Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2024/11/rpg-seminar.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20260520T110000
DTEND;TZID=Asia/Hong_Kong:20260520T120000
DTSTAMP:20260611T045558
CREATED:20260514T081134Z
LAST-MODIFIED:20260514T081134Z
UID:116062-1779274800-1779278400@ece.hku.hk
SUMMARY:RPG Seminar – Foundation-style Methods for Real-Time Statistical Dependency Measurement and Its Applications
DESCRIPTION:Zoom Link \nhttps://hku.zoom.us/j/9021481973?omn=94827890905 \nAbstract\nMutual information has long served as a principled measure of statistical dependence\, but computing it from empirical samples is notoriously difficult: neural estimators rely on costly gradient-based optimization for every new dataset\, which limits their use inside real-time and large-scale pipelines. This talk presents a framework that pretrains a neural estimator on a synthetic meta-distribution and then evaluates any new distribution in a single forward pass — turning a per-dataset optimization problem into an inference problem. Cheap\, online dependency readings further act as many applications\, including live training diagnostics\, and as a regularization signal for detecting key frames. \nSpeaker\nMr. Zhengyang HU\nDepartment of Electrical and Computer Engineering\nThe University of Hong Kong \nBiography of the Speaker\nZhengyang Hu\, a three-year PhD in ECE\, mainly focuses on statistical dependency measurement and foundational-style data science models. \nOrganiser\nProf Yanchao YANG \nDepartment of Electrical and Computer Engineering\, The University of Hong Kong \nAll are welcome.
URL:https://ece.hku.hk/events/20260520/
CATEGORIES:Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2024/11/rpg-seminar.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20260520T143000
DTEND;TZID=Asia/Hong_Kong:20260520T153000
DTSTAMP:20260611T045558
CREATED:20260518T064113Z
LAST-MODIFIED:20260518T064113Z
UID:116204-1779287400-1779291000@ece.hku.hk
SUMMARY:RPG Seminar – Dual Gate Phototransistors: Mechanism\, Performance Optimization\, and Application in SWIR Detection
DESCRIPTION:Zoom Link \nhttps://hku.zoom.us/j/8795454775?omn=92351306157 \nAbstract\nDual‑gate phototransistors enable independent control of carrier accumulation and separation within a single channel\, offering high gain\, low noise\, and tunable responsivity. To start\, this seminar reviews their fundamental mechanisms\, optimization strategies\, application in SWIR photodetection and novel applications. We then present a graphene/h‑BN/Se0.3Te0.7 dual‑gate phototransistor for short‑wave infrared (SWIR) detection. Graphene serves as top gate electrode\, h-BN is the top gate dielectric and Se0.3Te0.7 serves as the p-type channel. Electrical measurements reveal p‑type behavior\, with optimal on‑state current achieved under negative dual‑gate bias. Surprisingly\, under 1550 nm SWIR illumination\, the highest photoresponse occurs under positive dual‑gate bias. This opposite polarity dependence indicates that photoconductive gain is dominated by efficient collection of photogenerated carriers. The h‑BN layer suppresses dark current while providing a strong vertical field. Our results demonstrate that this heterostructure\, combined with dual‑gate modulation\, offers a scalable route for high‑performance\, low‑cost SWIR photodetection. \nSpeaker\nMr ZHOU Bufan\nDepartment of Electrical and Computer Engineering\nThe University of Hong Kong \nBiography of the Speaker\nBufan Zhou is an Mphil student in the Department of Electrical and Computer Engineering at the University of Hong Kong\, supervised by Prof. Can Li. He received his BSc. degree in Engineering Physics from the Hong Kong Polytechnic University in 2024. His research interests include 2D dual-gate phototransistors. \nOrganiser\nProf. Can LI \nDepartment of Electrical and Computer Engineering\, The University of Hong Kong \nAll are welcome.
URL:https://ece.hku.hk/events/20260520-2/
CATEGORIES:Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2024/11/rpg-seminar.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20260521T160000
DTEND;TZID=Asia/Hong_Kong:20260521T170000
DTSTAMP:20260611T045558
CREATED:20260420T064633Z
LAST-MODIFIED:20260515T085848Z
UID:115720-1779379200-1779382800@ece.hku.hk
SUMMARY:RPG Seminar – From Understanding to Intervention: Interpretability-Guided Methods for Improving Large Language Models
DESCRIPTION:  \nAbstract\nLarge language models have achieved impressive performance\, but improving them efficiently and reliably requires more than scaling alone. In this talk\, I present a series of works that explore how internal understanding of LLMs can be translated into practical interventions for better capability\, efficiency\, and controllability. I begin with actionable mechanistic interpretability\, introducing a unified “Locate\, Steer\, and Improve” perspective that turns model analysis into a framework for intervention. I then show how this perspective supports several concrete advances: data-free mixed-precision quantization guided by numerical and structural sensitivity\, multilingual capability enhancement through representation shifting and contrastive alignment\, personalized multi-teacher distillation that routes each prompt to its most suitable teacher\, and coarse-to-fine selective fine-tuning for mitigating catastrophic forgetting while preserving general versatility. Together\, these works reflect a common theme: interpretability is not only a tool for explaining LLMs\, but also a principled basis for designing more efficient training\, compression\, and adaptation methods. \nSpeaker\nMr Hengyuan ZHANG\nDepartment of Electrical and Computer Engineering\nThe University of Hong Kong \nBiography of the Speaker\nHengyuan Zhang is a Ph.D. candidate at the University of Hong Kong\, supervised by Prof. Ngai Wong and Prof. Hayden Kwok-Hay So. His research focuses on the improvement of efficiency and interpretability within large language models. He aims to uncover and characterize the internal processes that govern model behavior\, with the goal of improving model speciality\, interpretability\, and reliability in real-world deployments. He has published multiple papers in leading venues such as ACL\, EMNLP\, TKDD\, and NeurIPS. \nOrganiser\nProf. Ngai WONG \nDepartment of Electrical and Computer Engineering\, The University of Hong Kong \nAll are welcome.
URL:https://ece.hku.hk/events/20260521-1/
LOCATION:Room CB-603\, 6/F\, Chow Yei Ching Building\, The University of Hong Kong
CATEGORIES:Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2024/11/rpg-seminar.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20260527T093000
DTEND;TZID=Asia/Hong_Kong:20260527T103000
DTSTAMP:20260611T045558
CREATED:20260521T022028Z
LAST-MODIFIED:20260521T022028Z
UID:116420-1779874200-1779877800@ece.hku.hk
SUMMARY:RPG Seminar – Can Image Models Think? Benchmarking and Empowering Models with Knowledge and Reasoning
DESCRIPTION:Zoom Link \nhttps://hku.zoom.us/meetings/95832932074/invitations?signature=SCg4igomh110T7WUpW6op2xeFgb9WgU92xDTYI2sp8s \nAbstract\nText-to-image generation models have achieved impressive visual quality\, yet they largely fail when prompts require reasoning or world knowledge to interpret. Generating an image of “a park thirty minutes after heavy rainfall” or “the city hosting the 2021 Summer Olympics” demands inference and domain knowledge\, not just pattern matching. This seminar examines the gap between what current T2I models can generate and what they can genuinely understand\, exploring how reasoning can be evaluated\, measured\, and ultimately integrated into the image generation process. \nSpeaker\nMiss Kaiyue SUN\nDepartment of Electrical and Computer Engineering\nThe University of Hong Kong \nBiography of the Speaker\nI am a PhD student at the Department of Electrical and Computer Engineering\, University of Hong Kong\, supervised by Prof. Xihui Liu. Before joining HKU\, I received my bachelor’s degree in Electrical and Electronic Engineering at Imperial College London. My research interests cover deep learning and artificial intelligence\, with special emphasis on generative models and unified multimodal models. My publications have been received by NeurIPS\, CVPR\, ACL\, and T-PAMI. \nOrganiser\nProf. Xihui LIU \nDepartment of Electrical and Computer Engineering\, The University of Hong Kong \nAll are welcome.
URL:https://ece.hku.hk/events/20260527-2/
CATEGORIES:Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2024/11/rpg-seminar.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20260527T103000
DTEND;TZID=Asia/Hong_Kong:20260527T110000
DTSTAMP:20260611T045558
CREATED:20260521T020149Z
LAST-MODIFIED:20260521T020149Z
UID:116417-1779877800-1779879600@ece.hku.hk
SUMMARY:RPG Seminar – Speculative Jacobi Decoding: Pushing the Limits of Parallelism in Autoregressive Visual Generation
DESCRIPTION:Zoom Link \nhttps://hku.zoom.us/s/92652052314?pwd=kKKw9mhMHxP8xcObYHFop7VyvAfzh1.1#success \nAbstract\nAutoregressive (AR) models have emerged as a powerful paradigm for text-to-image and text-to-video generation\, exhibiting exceptional native in-context learning capabilities and scalable performance. However\, their real-world deployment remains severely bottlenecked by the conventional sequential\, token-by-token decoding process\, which incurs immense inference latency by demanding thousands of sequential forward passes. To break this fundamental sequential dependency\, this talk introduces the Speculative Jacobi Decoding (SJD) framework. SJD reformulates the sequential prediction of AR models as a system of non-linear equations\, enabling multi-token parallel decoding through a novel probabilistic acceptance criterion that preserves the critical visual diversity required in sampling-based generation. Without requiring any auxiliary networks or model fine-tuning\, SJD achieves significant\, training-free wall-clock speedups across standard visual benchmarks while maintaining identical generation quality\, bridging the gap between powerful AR architectures and efficient real-world deployment. \nSpeaker\nMr Yao TENG\nDepartment of Electrical and Computer Engineering\nThe University of Hong Kong \nBiography of the Speaker\nYao Teng is currently a PhD student in the Department of Electrical and Electronic Engineering at The University of Hong Kong (HKU)\, under the supervision of Dr. Xihui Liu. His research interests lie broadly in computer vision\, deep generative models\, and efficient machine learning inference. \nOrganiser\nProf Xihui LIU \nDepartment of Electrical and Computer Engineering\, The University of Hong Kong \nAll are welcome.
URL:https://ece.hku.hk/events/20260527-1/
CATEGORIES:Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2024/11/rpg-seminar.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20260527T140000
DTEND;TZID=Asia/Hong_Kong:20260527T150000
DTSTAMP:20260611T045558
CREATED:20260521T090701Z
LAST-MODIFIED:20260521T090701Z
UID:116423-1779890400-1779894000@ece.hku.hk
SUMMARY:RPG Seminar – Multi-Dimensional Nano-Printing of Colloidal Quantum Dots for Infrared Optoelectronics
DESCRIPTION:Zoom Link \nhttps://hku.zoom.us/j/94632758376 \nAbstract\nAdditive manufacturing enables customised device fabrication for emerging sensing technologies.  However\, printable (opto)electronic devices with sophisticated architectures\, including all-printed photodiodes\, face challenges in multi-material and multi-layer printing at micro- and nanoscales with low processing temperatures.  Herein\, we establish a nano-resolution printing method based on electrohydrodynamic printing (EHDP) to deposit inks from the colloidal nanocrystal (NC) library\, followed by in situ room-temperature ligand exchange to functionalise the NC solids.  This general approach enables layer-by-layer printing with wide selections of NC inks\, ligand reagents\, substrates\, and device architectures.  Chemical-treatment-induced contraction and densification grants printed Ag NC structures electrical conductivity and an achievable feature size and filling ratio of 70 nm and 75%\, respectively\, constructing wide-gamut structural colour gratings.  By exploiting Ag\, Au\, PbS\, and ZnO NCs and compact ligands\, we demonstrate all-printed multi-layer infrared photodiodes with sub-10-µm pixel sizes.  The nano-printing assembly of hetero-NCs promises the facile integration of multi-functional micro-nano devices. \nSpeaker\nMr. Zhixuan ZHAO \nDepartment of Electrical and Computer Engineering \nThe University of Hong Kong \nBiography of the Speaker\nZhixuan Zhao is a fourth-year Ph.D. student in the Department of Electrical and Computer Engineering at the The University of Hong Kong (HKU) under the guidance of Prof. Leo Tianshuo Zhao. He earned a master’s degree in mechanical manufacturing and automation from Northwestern Polytechnical University and a bachelor’s degree from Hefei University of Technology. \nOrganiser\nProf. Leo Tianshuo ZHAO \nDepartment of Electrical and Computer Engineering\, The University of Hong Kong \nAll are welcome.
URL:https://ece.hku.hk/events/20260527-3/
CATEGORIES:Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2024/11/rpg-seminar.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20260528T100000
DTEND;TZID=Asia/Hong_Kong:20260528T110000
DTSTAMP:20260611T045558
CREATED:20260522T025938Z
LAST-MODIFIED:20260522T025938Z
UID:116482-1779962400-1779966000@ece.hku.hk
SUMMARY:RPG Seminar – Controllable 3D Content Generation
DESCRIPTION:Zoom Link \nhttps://hku.zoom.us/j/3623268282 \nAbstract\nCreating high-quality 3D content requires not only visual realism but also controllability\, editability\, and structural understanding. This talk focuses on recent progress toward controllable 3D content generation\, tracing a path from holistic object-level generation to part-aware modeling. It begins with object-level controllable 3D generation enabled by multi-view conditioning\, aiming to guide the overall shape\, appearance\, and structure of generated objects. The talk then moves toward more fine-grained control\, emphasizing the importance of decomposing 3D objects into meaningful parts for downstream applications. Finally\, it highlights how part-level representations can support broader goals in 3D generation and embodied AI. \nSpeaker\nMr Yunhan YANG \nDepartment of Electrical and Computer Engineering \nThe University of Hong Kong \nBiography of the Speaker\nYunhan Yang is currently a PhD student in the Department of Electrical and Electronic Engineering at The University of Hong Kong\, under the supervision of Dr. Xihui Liu. His research interests lie broadly in computer vision\, with special emphasis on 3D vision. \nOrganiser\nProf Xihui LIU \nDepartment of Electrical and Computer Engineering\, The University of Hong Kong \nAll are welcome.
URL:https://ece.hku.hk/events/20260528-1/
CATEGORIES:Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2024/11/rpg-seminar.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20260529T100000
DTEND;TZID=Asia/Hong_Kong:20260529T110000
DTSTAMP:20260611T045558
CREATED:20260526T020314Z
LAST-MODIFIED:20260526T020314Z
UID:116740-1780048800-1780052400@ece.hku.hk
SUMMARY:RPG Seminar – Embodied Spatial Intelligence for Generalizable Robot Navigation
DESCRIPTION:Zoom Link \nhttps://hku.zoom.us/j/97268323750?pwd=rqO1zHBqvsVUD01hj31lRSV9r6Adhp.1 \nAbstract\nBuilding embodied agents with spatial intelligence remains a fundamental challenge in robotics and embodied AI. Such agents must understand open-vocabulary semantics\, maintain long-term spatial memory\, and perform robust real-time navigation in dynamic environments. This talk presents a progression toward generalizable embodied spatial intelligence\, spanning open-vocabulary exploration\, streaming vision-language navigation\, and dual-system navigation architectures that unify high-level reasoning with real-time control. \nSpeaker\nMiss Meng WEI \nDepartment of Electrical and Computer Engineering \nThe University of Hong Kong \nBiography of the Speaker\nMeng Wei is currently a PhD student in the Department of Electrical and Electronic Engineering at The University of Hong Kong\, under the supervision of Dr. Xihui Liu. His research interests lie broadly in embodied AI and robot learning\, with a particular focus on vision-language navigation\, embodied spatial intelligence\, and foundation models for long-horizon robotic reasoning and control. \nOrganiser\nProf Xihui LIU \nDepartment of Electrical and Computer Engineering\, The University of Hong Kong \nAll are welcome.
URL:https://ece.hku.hk/events/20260529-1/
CATEGORIES:Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2024/11/rpg-seminar.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20260602T103000
DTEND;TZID=Asia/Hong_Kong:20260602T113000
DTSTAMP:20260611T045558
CREATED:20260529T100852Z
LAST-MODIFIED:20260601T084435Z
UID:117024-1780396200-1780399800@ece.hku.hk
SUMMARY:Seminar on Physics-Aware World Model for Video Generation and Embodied AI
DESCRIPTION:The seminar on “Physics-Aware World Model for Video Generation and Embodied AI” is rescheduled to begin at 10:30 am. \nAbstract\nIn AI and cognitive science\, world models are key for planning\, reasoning\, and learning from experience. An effective world model needs to: sense and learn real-world knowledge\, predict and generate real-world scenes\, reason and control according to physical laws\, and act robustly with human-in-the-loop. Prior work on world models has limited capability in representation/generation and physical awareness. We overcome these limitations through two innovations\, and towards the first open-source\, physically grounded world model from academia. First\, we develop a flow matching and DPO reinforcement learning framework to improve the continuity and physical awareness in world model representation and generation. Our world model PhyWorld achieves simultaneously best-of-the-results in physical awareness and state-of-the-art in open-source video generation. Second\, we develop a comprehensive physical awareness benchmarking and arena system. We extract comprehensively 50-60 metrics demonstrating physical law awareness of video generation and trajectory. Our comprehensive benchmark PhyGround includes video quality\, common sense\, Newtonian mechanics\, optics\, energy\, chemical\, materials\, etc.\, which is missing in literature. We generate an agent and a 9B language model for evaluating physical awareness according to these benchmarks. Last\, we describe our effort and results towards the concept of “world model for all”\, which utilises a single world model for robotic control and robot navigation\, task management\, planning\, task decomposition for high-level management and control\, as well as automatic SLAM and 3D reconstruction for environment sensing. \nSpeaker\nProf. Yanzhi WANG\nProfessor\,\nDepartment of Electrical and Computer Engineering\,\nNortheastern University \nSpeaker’s Biography\nYanzhi WANG is a Professor in the Department of Electrical and Computer Engineering and Computer Science at Northeastern University\, a senior member of IEEE. His research interests focus on real-time and energy-efficient deep learning and artificial intelligence systems\, especially on efficient large language models and large-scale generative AI systems. His research works have been published broadly in (i) machine learning conferences such as AAAI\, CVPR\, NeurIPS\, ICML\, ICCV\, ICLR\, IJCAI\, ECCV\, KDD\, ICRA\, ACM MM\, ICDM\, etc.\, (ii) architecture and system conferences such as ASPLOS\, ISCA\, MICRO\, HPCA\, CCS\, VLDB\, PLDI\, WWW\, ICS\, PACT\, CGO\, IPDPS\, INFOCOM\, ICDCS\, DAC\, ICCAD\, FPGA\, FCCM\, ISSCC\, CICC\, RTAS\, RTSS\, etc.\, and (iii) IEEE and ACM transactions. His research works have been cited 29\,000 times. He has received six Best Paper Awards and another 12 Best Paper Nominations. He has many research awards\, including the U.S. Army Young Investigator Award and other young investigator awards\, research awards from Google\, Intel\, Mathworks\, etc. His research work has been reported and cited by around 500 media. He has 18 academic descendants as tenure-track faculty members at the University of Minnesota\, University of Massachusetts Amherst\, University of Arizona\, City University of Hong Kong\, University of Georgia\, Zhejiang University\, etc. His alumni have become key contributors in Google Gemini and Alibaba Qwen 3. \nOrganiser\nProf. Kaibin HUANG\nDepartment of Electrical and Computer Engineering\,\nThe University of Hong Kong\n\nAll are welcome!
URL:https://ece.hku.hk/events/20260602-1/
LOCATION:Room CB-603\, 6/F\, Chow Yei Ching Building\, The University of Hong Kong
CATEGORIES:Highlights,Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2026/05/1280-1.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20260609T103000
DTEND;TZID=Asia/Hong_Kong:20260609T113000
DTSTAMP:20260611T045558
CREATED:20260601T100902Z
LAST-MODIFIED:20260601T100902Z
UID:117067-1781001000-1781004600@ece.hku.hk
SUMMARY:Seminar on AI-native 6G Architecture: Distributed Intelligence and Programmable Wireless Systems
DESCRIPTION:Abstract\nThis keynote examines the architectural transition from deterministic 5G design to AI-native 6G\, where intelligence is embedded across device\, RAN\, core\, operations\, and lifecycle layers to meet extreme targets in throughput\, latency\, positioning\, and energy efficiency. Rather than treating AI as an external optimisation layer\, the talk presents 6G as a distributed and federated system in which learned policies reshape core engineering\, radio control\, and spectrum management itself. The presentation then shows how this architectural shift brings together several structural enablers and consequences: network slicing and policy control in the core\, AI-assisted beamforming in the RAN\, RIS and NTN as programmable extensions of wireless coverage\, ISAC as a sensing-capable infrastructure layer\, and ONNX/Open RAN style interoperability as part of deployable openness. It also addresses the associated constraints\, including inference-driven energy costs\, the need for explainable AI and auditability\, outcome-based licensing for dynamic spectrum access\, and the operational challenge of post-quantum security at telecom scale. The keynote argues that 6G should be understood not as a faster air interface\, but as an integrated\, programmable\, and governable cyber-physical network system whose value and risk emerge from the interaction of its components. \nSpeaker\nProf. Soumaya CHERKAOUI\nPolytechnique Montréal\, Canada \nSpeaker’s Biography\nProf. Soumaya CHERKAOUI is a Full Professor at the Department of Computer and Software Engineering at Polytechnique Montréal\, Canada. Her research interests are in wireless networks. She specialises in the convergence of machine learning and communication networks\, as well as the application of quantum technologies in future communication systems. Before joining academia as a Professor in 1999\, she worked in the industry as a project leader on aerospace-related projects. Pr. Cherkaoui has published over 200 research papers in renowned journals and conferences. She has served as a guest editor and an editorial board member for several prestigious journals\, including IEEE Journal on Selected Areas in Communications (JSAC)\, IEEE Network\, IEEE Systems\, and Computer Networks (Wiley and Elsevier). Her research has led to successful technology transfers and has been recognised with multiple accolades\, including best paper awards\, most notably at the IEEE Communications Society’s flagship conference\, IEEE ICC 2017\, with recent awards at ICCSPA 2024\, GIIS 2025\, and IEEE LCN 2024. She has played key leadership roles in major conferences\, serving as the General Chair of IEEE LCN 2019\, Technical Co-Chair of IEEE ICC 2025\, and Symposium Co-Chair for IEEE ICC 2018\, IEEE Globecom 2018\, IEEE Globecom 2015\, and IEEE ICC 2014. Additionally\, she was Chair of the IEEE Communications Society’s Technical Committee on IoT\, Ad Hoc\, and Sensor Networks (2020–2021)\, the second-largest technical committee within IEEE ComSoc. Prof. Cherkaoui is the recipient of the C.C. Gotlieb Computer Award\, the Mirela Notere Award and the IEEE Bio-Inspired Computing STC Leadership Award and was recognised as a “Star” in Networking and Communications by N2Women. She is also an IEEE Communications Society Distinguished Lecturer. \nOrganiser\nProf. Edith Cheuk Han NGAI\nDepartment of Electrical and Computer Engineering\,\nThe University of Hong Kong \nSupported By \nTam Wing Fan Innovation Wing Two\n\nAll are welcome!
URL:https://ece.hku.hk/events/20260609-1/
LOCATION:Tam Wing Fan Innovation Wing Two\, G/F\, Run Run Shaw Building\, The University of Hong Kong
CATEGORIES:Highlights,Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2026/06/1280-1.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20260617T110000
DTEND;TZID=Asia/Hong_Kong:20260617T120000
DTSTAMP:20260611T045558
CREATED:20260609T085437Z
LAST-MODIFIED:20260609T085437Z
UID:117126-1781694000-1781697600@ece.hku.hk
SUMMARY:RPG Seminar – Adaptive Lightweight Learning for Edge Non-Intrusive Load Monitoring
DESCRIPTION:Zoom Link \nhttps://us05web.zoom.us/j/88334748156?pwd=YXWVQUg4FEHYb9nucvsm78fqasXabh.1 \nAbstract\nNon-intrusive load monitoring provides appliance-level electricity consumption information from aggregate household measurements\, supporting residential energy management without installing sensors on individual appliances. Processing these measurements on local edge devices can provide timely feedback while preserving user privacy. However\, practical deployment remains difficult because edge devices have limited computing and memory resources\, and differences among households can reduce the accuracy of models trained elsewhere. This seminar presents a lightweight learning framework that enables non-intrusive load monitoring models to operate on resource-constrained devices and adapt to previously unseen households without requiring appliance-level labels from local users. The study examines the full process from preparing a model for edge deployment to improving its performance after deployment. Ultimately\, the framework enables accurate\, low-cost\, and privacy-preserving household energy monitoring across diverse edge devices and residential environments. \nSpeaker\nMr Taoyu LU \nDepartment of Electrical and Computer Engineering \nThe University of Hong Kong \nBiography of the Speaker\nTaoyu LU received the B.Eng. degree in electrical engineering and automation from Huazhong University of Science and Technology in 2024. He is currently pursuing the M.Phil. degree in electrical and electronic engineering at the University of Hong Kong. His current research interests include low-carbon energy systems and edge intelligence in smart grids. \nOrganiser\nProf. Yi WANG \nDepartment of Electrical and Computer Engineering\, The University of Hong Kong \nAll are welcome.
URL:https://ece.hku.hk/events/20260617-1/
CATEGORIES:Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2024/11/rpg-seminar.jpg
END:VEVENT
END:VCALENDAR