Difference between revisions of "Resource:Previous Seminars"

From MobiNetS
Jump to: navigation, search
 
(6 intermediate revisions by 3 users not shown)
Line 1: Line 1:
=== History ===
=== History ===
{{Hist_seminar
|abstract = Reconfigurable Intelligent Surfaces (RIS) are a promising technology for creating smart radio environments by controlling wireless propagation. However, several factors hinder the integration of RIS technology into existing cellular networks, including the incompatibility of RIS control interfaces with 5G PHY/MAC procedures for synchronizing radio scheduling decisions and RIS operation, and the cost and energy limitations of passive RIS technology. This paper presents RISENSE, a system for practical RIS integration in cellular networks. First, we propose a novel, low-cost, and low-power RIS design capable of decoding control messages without complex baseband operations or additional RF chains, utilizing a power sensor and a network of microstrip lines and couplers. Second, we design an effective in-band wireless RIS control interface, compatible with 5G PHY/MAC procedures, that embeds amplitude-modulated (AM) RIS control commands directly into standard OFDM-modulated 5G data channels. Finally, we propose a low-overhead protocol that supports swift on-demand RIS re-con gurability, making it adaptable to varying channel conditions and user mobility, while minimizing the wastage of 5G OFDM symbols. Our experiments validate the design of RISENSE and our evaluation shows that our system can reconfigure a RIS at the same pace as users move, boosting 5G coverage where static or slow RIS controllers cannot.
|confname = Mobisys'25
|link = https://dspace.networks.imdea.org/handle/20.500.12761/1925
|title= RISENSE: Long-Range In-Band Wireless Control of Passive Reconfigurable Intelligent Surfaces
|speaker= Haifeng
|date=2025-9-12
}}
{{Hist_seminar
|abstract = Traditional 3D content representations include dense point clouds that consume large amounts of data and hence network bandwidth, while newer representations such as neural radiance fields suffer from poor frame rates due to their non-standard volumetric rendering pipeline. 3D Gaussian splats (3DGS) can be seen as a generalization of point clouds that meet the best of both worlds, with high visual quality and efficient rendering for real-time frame rates. However, delivering 3DGS scenes from a hosting server to client devices is still challenging due to high network data consumption (e.g., 1.5 GB for a single scene). The goal of this work is to create an efficient 3D content delivery framework that allows users to view high quality 3D scenes with 3DGS as the underlying data representation. The main contributions of the paper are: (1) Creating new layered 3DGS scenes for efficient delivery, (2) Scheduling algorithms to choose what splats to download at what time, and (3) Trace-driven experiments from users wearing virtual reality headsets to evaluate the visual quality and latency. Our system for Layered 3D Gaussian Splats delivery (L3GS) demonstrates high visual quality, achieving 16.9% higher average SSIM compared to baselines, and also works with other compressed 3DGS representations. The code is available at https://github.com/mavens-lab/layered_3d_gaussian_splats.
|confname =Mobicom'25
|link = https://arxiv.org/html/2504.05517v1
|title= L3GS: Layered 3D Gaussian Splats for Efficient 3D Scene Delivery
|speaker=Jiyi
|date=2025-9-12
}}
{{Hist_seminar
|abstract = This year, we are embracing the exciting new trends in AIoT including MLsys, LLMs, embodied perception, volumetric videos, etc. Papers collected from top venues in 2025 will be discussed in-depth, and research problems and new ideas are to be discovered!
|confname = Begin of new semester
|link = https://mobinets.cn/site/Resource:Paper_Carnival_2025
|title= Paper Carnival 2025
|speaker=All
|date=2025-08-27
}}
{{Hist_seminar
|abstract = In the metaverse era, point cloud video (PCV) streaming on mobile XR devices is pivotal. While most current methods focus on PCV compression from traditional 3-DoF video services, emerging AI techniques extract vital semantic information, producing content resembling the original. However, these are early-stage and computationally intensive. To enhance the inference efficacy of AI-based approaches, accommodate dynamic environments, and facilitate applicability to metaverse XR devices, we present ISCom, an interest-aware semantic communication scheme for lightweight PCV streaming. ISCom is featured with a region-of-interest (ROI) selection module, a lightweight encoder-decoder training module, and a learning-based scheduler to achieve real-time PCV decoding and rendering on resource-constrained devices. ISCom’s dual-stage ROI selection provides significantly reduces data volume according to real-time interest. The lightweight PCV encoder-decoder training is tailored to resource-constrained devices and adapts to the heterogeneous computing capabilities of devices. Furthermore, We provide a deep reinforcement learning (DRL)-based scheduler to select optimal encoder-decoder model for various devices adaptivelly, considering the dynamic network environments and device computing capabilities. Our extensive experiments demonstrate that ISCom outperforms baselines on mobile devices, achieving a minimum rendering frame rate improvement of 10 FPS and up to 22 FPS. Furthermore, our method significantly reduces memory usage by 41.7% compared to the state-of-the-art AITransfer method. These results highlight the effectiveness of ISCom in enabling lightweight PCV streaming and its potential to improve immersive experiences for emerging metaverse application.
|confname =JSAC'24
|link = https://dl.acm.org/doi/10.1109/JSAC.2023.3345430
|title= ISCom: Interest-Aware Semantic Communication Scheme for Point Cloud Video Streaming on Metaverse XR Devices
|speaker=Jiyi
|date=2025-06-13
}}
{{Hist_seminar
|abstract = Scientific Illustration Tutorial
|confname = TUTORIAL
|link = https://mobinets.cn/Resource:Seminar
|title= Idea share
|speaker=OldBee
|date=2025-06-13
}}
{{Hist_seminar
|abstract = Deploying deep convolutional neural networks (CNNs) for edge-based video analytics poses significant challenges due to the intensive computing demands. Model partitioning has emerged as a promising solution by offloading segments of CNNs to multiple proximal edge devices for collaborative inference. However, this approach often incurs substantial cross-device transmission overhead, particularly in handling intermediate feature maps. To address these limitations, we propose ReDream (REsidual feature-DRivEn mixed spArse coding for Model partitioning), a novel edge-centric video analytics framework that jointly optimizes  transmission efficiency and inference accuracy. ReDream introduces two key innovations: 1) It enhances the sparsity of intermediate features by replacing activation functions with ReLU in selected CNN layers and retraining, thereby increasing the proportion of zero-valued elements. 2) It leverages the heterogeneous distribution of feature data across layers by applying a mixed sparse coding scheme, i.e., selecting different compression methods adaptively to optimize model partitioning. These optimizations enable ReDream to support more efficient cross-device inference while maintaining high model accuracy, making it well-suited for real-time deployment in collaborative edge environments.
|confname = IDEA
|link = https://mns.uestc.cn/wiki/Research:InProgress/MixedSparseCoding
|title= ReDream: Residual Feature-Driven Mixed Sparse Coding for Model Partitioning
|speaker=Xianyang
|date=2025-05-23
}}
{{Hist_seminar
|abstract = While existing strategies to execute deep learning-based classification on low-power platforms assume the models are trained on all classes of interest, this paper posits that adopting context-awareness i.e. narrowing down a classification task to the current deployment context consisting of only recent inference queries can substantially enhance performance in resource-constrained environments. We propose a new paradigm, CACTUS, for scalable and efficient context-aware classification where a micro-classifier recognizes a small set of classes relevant to the current context and, when context change happens (e.g., a new class comes into the scene), rapidly switches to another suitable micro-classifier. CACTUS features several innovations, including optimizing the training cost of context-aware classifiers, enabling on-the-fly context-aware switching between classifiers, and balancing context switching costs and performance gains via simple yet effective switching policies. We show that CACTUS achieves significant benefits in accuracy, latency, and compute budget across a range of datasets and IoT platforms.
|confname = Mobisys'24
|link = https://dl.acm.org/doi/abs/10.1145/3643832.3661888
|title= CACTUS: Dynamically Switchable Context-aware micro-Classifiers for Efficient IoT Inference
|speaker= Zhenhua
|date=2025-04-18
}}
{{Hist_seminar
|abstract = Nowadays, volumetric videos have emerged as an attractive multimedia application providing highly immersive watching experiences since viewers could adjust their viewports at 6 degrees-of-freedom. However, the point cloud frames composing the video are prohibitively large, and effective compression techniques should be developed. There are two classes of compression methods. One suggests exploiting the conventional video codecs (2D-based methods) and the other proposes to compress the points in 3D space directly (3D-based methods). Though the 3D-based methods feature fast coding speeds, their compression ratios are low since the failure of leveraging inter-frame redundancy. To resolve this problem, we design a patch-wise compression framework working in the 3D space. Specifically, we search rigid moves of patches via the iterative closest point algorithm and construct a common geometric structure, which is followed by color compensation. We implement our decoder on a GPU platform so that real-time decoding and rendering are realized. We compare our method with GROOT, the state-of-the-art 3D-based compression method, and it reduces the bitrate by up to 5.98×. Moreover, by trimming invisible content, our scheme achieves comparable bandwidth demand of V-PCC, the representative 2D-based method, in FoV-adaptive streaming.
|confname = TC'24
|link = https://ieeexplore.ieee.org/document/10360355
|title= A GPU-Enabled Real-Time Framework for Compressing and Rendering Volumetric Videos
|speaker=Mengfan
|date=2025-04-18
}}
{{Hist_seminar
|abstract = Cross-silo federated learning (FL) enables multiple institutions (clients) to collaboratively build a global model without sharing their private data. To prevent privacy leakage during aggregation, homomorphic encryption (HE) is widely used to encrypt model updates, yet incurs high computation and communication overheads. To reduce these overheads, packed HE (PHE) has been proposed to encrypt multiple plaintexts into a single ciphertext. However, the original design of PHE does not consider the heterogeneity among different clients, an intrinsic problem in cross-silo FL, often resulting in undermined training efficiency with slow convergence and stragglers. In this work, we propose FedPHE, an efficiently packed homomorphically encrypted FL framework with secure weighted aggregation and client selection to tackle the heterogeneity problem. Specifically, using CKKS with sparsification, FedPHE can achieve efficient encrypted weighted aggregation by accounting for contributions of local updates to the global model. To mitigate the straggler effect, we devise a sketching-based client selection scheme to cherry-pick representative clients with heterogeneous models and computing capabilities. We show, through rigorous security analysis and extensive experiments, that FedPHE can efficiently safeguard clients’ privacy, achieve a training speedup of 1.85 − 4.44×, cut the communication overhead by 1.24 − 22.62× , and reduce the straggler effect by up to 1.71 − 2.39×.
|confname =INFOCOM24'
|link = https://ieeexplore.ieee.org/abstract/document/10621440
|title= Efficient and Straggler-Resistant Homomorphic Encryption for Heterogeneous Federated Learning
|speaker=Dongting
|date=2025-03-28
}}{{Hist_seminar
|abstract = Entanglement routing (ER) in quantum networks must guarantee entanglement fidelity, a property that is crucial for applications such as quantum key distribution, quantum computation, and quantum sensing. Conventional ER approaches assume that network links can only generate entanglements with a fixed fidelity, and then they rely on purification to improve endto-end fidelities. However, recent advances in entanglement generation technologies show that quantum links can be configured by choosing among different fidelity/entanglement-rate combinations (defined in this paper as link configurations), hence enabling a more flexible assignment of quantum-network resources for meeting specific application requirements. To exploit this opportunity, we introduce the problem of link configuration for fidelityconstrained routing and purification (LC-FCRP) in Quantum Networks. We first formulate a simplified FCRP version as a Mixed Integer Linear Programming (MILP) model, where the link fidelity can be adjusted within a finite set. Then, to explore the full space of possible link configurations, we propose a link configuration algorithm based on a novel shortest-pathbased fidelity determination (SPFD) algorithm w/o Bayesian Optimization, which can be applied on top of any existing ER algorithm. Numerical results demonstrate that link configuration improves the acceptance ratio of existing ER algorithms by 87%.
|confname =INFOCOM25'
|link = https://re.public.polimi.it/bitstream/11311/1281986/1/final_infocom25_link_configuration_for_entanglement_routing.pdf
|title= Link Configuration for Fidelity-Constrained Entanglement Routing in Quantum Networks
|speaker=Yaliang
|date=2025-03-27
}}
{{Hist_seminar
{{Hist_seminar
|abstract = Large language models (LLMs) have demonstrated remarkable reasoning capabilities across diverse domains. Recent studies have shown that increasing test-time computation enhances LLMs' reasoning capabilities. This typically involves extensive sampling at inference time guided by an external LLM verifier, resulting in a two-player system. Despite external guidance, the effectiveness of this system demonstrates the potential of a single LLM to tackle complex tasks. Thus, we pose a new research problem: Can we internalize the searching capabilities to fundamentally enhance the reasoning abilities of a single LLM? This work explores an orthogonal direction focusing on post-training LLMs for autoregressive searching (i.e., an extended reasoning process with self-reflection and self-exploration of new strategies). To achieve this, we propose the Chain-of-Action-Thought (COAT) reasoning and a two-stage training paradigm: 1) a small-scale format tuning stage to internalize the COAT reasoning format and 2) a large-scale self-improvement stage leveraging reinforcement learning. Our approach results in Satori, a 7B LLM trained on open-source models and data. Extensive empirical evaluations demonstrate that Satori achieves state-of-the-art performance on mathematical reasoning benchmarks while exhibits strong generalization to out-of-domain tasks. Code, data, and models will be fully open-sourced.
|abstract = Large language models (LLMs) have demonstrated remarkable reasoning capabilities across diverse domains. Recent studies have shown that increasing test-time computation enhances LLMs' reasoning capabilities. This typically involves extensive sampling at inference time guided by an external LLM verifier, resulting in a two-player system. Despite external guidance, the effectiveness of this system demonstrates the potential of a single LLM to tackle complex tasks. Thus, we pose a new research problem: Can we internalize the searching capabilities to fundamentally enhance the reasoning abilities of a single LLM? This work explores an orthogonal direction focusing on post-training LLMs for autoregressive searching (i.e., an extended reasoning process with self-reflection and self-exploration of new strategies). To achieve this, we propose the Chain-of-Action-Thought (COAT) reasoning and a two-stage training paradigm: 1) a small-scale format tuning stage to internalize the COAT reasoning format and 2) a large-scale self-improvement stage leveraging reinforcement learning. Our approach results in Satori, a 7B LLM trained on open-source models and data. Extensive empirical evaluations demonstrate that Satori achieves state-of-the-art performance on mathematical reasoning benchmarks while exhibits strong generalization to out-of-domain tasks. Code, data, and models will be fully open-sourced.

Latest revision as of 18:49, 16 September 2025

History

2024

2023

2022

2021

2020

  • [Topic] [ The path planning algorithm for multiple mobile edge servers in EdgeGO], Rong Cong, 2020-11-18

2019

2018

2017

Instructions

请使用Latest_seminar和Hist_seminar模板更新本页信息.

    • 修改时间和地点信息
    • 将当前latest seminar部分的code复制到这个页面中
    • 将{{Latest_seminar... 修改为 {{Hist_seminar...，并增加对应的日期信息|date=
    • 填入latest seminar各字段信息
    • link请务必不要留空，如果没有link则填本页地址 https://mobinets.org/index.php?title=Resource:Seminar
  • 格式说明
    • Latest_seminar:

{{Latest_seminar
|confname=
|link=
|title=
|speaker=
}}

    • Hist_seminar

{{Hist_seminar
|confname=
|link=
|title=
|speaker=
|date=
}}

Retrieved from "http://mobinets.cn/site/index.php?title=Resource:Previous_Seminars&oldid=3423"