Publications
publications by categories in reversed chronological order. generated by jekyll-scholar.
2026
- TokenFlow: Responsive LLM Text Streaming Serving under Request Burst via Preemptive SchedulingIn Proceedings of the 21st European Conference on Computer Systems (EuroSys), 2026
2025
- MURAL: A Multi-Resolution Anytime Framework for LiDAR Object Detection Deep Neural NetworksIn 2025 IEEE 31st International Conference on Embedded and Real-Time Computing Systems and Applications (RTCSA), 2025
- DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN TrainingIn Proceedings of the 23rd ACM International Conference on Mobile Systems, Applications, and Services (MobiSys), 2025
- Better Reliability Compression: Model Pruning with Calibrated Uncertainty Estimation for Mobile Deep Learning ApplicationsIn 2025 IEEE 3rd International Conference on Mobility, Operations, Services and Technologies (MOST), 2025
- On-Device Dynamic DNN Inference through Spatial Sparsity ExploitationGetMobile: Mobile Computing and Communications, 2025
2024
- Taming algorithmic priority inversion in mission-critical perception pipelinesCommunications of the ACM, 2024
- VALO: a versatile anytime framework for LiDAR-based object detection deep neural networksIEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (also in EMSOFT), 2024
- DynaSpa: Exploiting Spatial Sparsity for Efficient Dynamic DNN Inference on DevicesIn Proceedings of the 22nd ACM Conference on Embedded Networked Sensor Systems (SenSys), 2024
2023
- A Unified Knowledge Distillation Framework for Deep Directed Graphical ModelsIn Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023
- Generalized self-cueing real-time attention scheduling with intermittent inspection and image resizingReal-Time Systems, 2023
- Scaleflow: Efficient deep vision pipeline with closed-loop scale-adaptive inferenceIn Proceedings of the 31st ACM International Conference on Multimedia, 2023
- Neural Network Models for Time Series DataIn Artificial Intelligence for Edge Computing, 2023
- Model compression for edge computingIn Artificial Intelligence for Edge Computing, 2023
2022
- Semi-supervised hypergraph node classification on hypergraph line expansionIn Proceedings of the 31st ACM international conference on information & knowledge management, 2022
- Self-cueing real-time attention scheduling in criticality-aware visual machine perceptionIn 2022 IEEE 28th Real-Time and Embedded Technology and Applications Symposium (RTAS), 2022
- Anytime-Lidar: Deadline-aware 3D object detectionIn 2022 IEEE 28th International Conference on Embedded and Real-Time Computing Systems and Applications (RTCSA), 2022
- Exploring spherical autoencoder for spherical video content processingIn MM’22: Proceedings of the 30th ACM International Conference on Multimedia, 2022
2021
- ControlVAE: Tuning, analytical properties, and performance analysisIEEE transactions on pattern analysis and machine intelligence, 2021
- CrossRoI: Cross-camera region of interest optimization for efficient real time video analytics at scaleIn Proceedings of the 12th ACM Multimedia Systems Conference, 2021
- Dydiff-vae: A dynamic variational framework for information diffusion predictionIn Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval, 2021
- Simulating Online Social Response: A Stimulus/Response PerspectiveIn 2021 Winter Simulation Conference (WSC), 2021
- Deep compressive offloading: Speeding up edge offloading for AI servicesGetMobile: Mobile Computing and Communications, 2021
- On orthogonality constraints for transformersIn Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing, 2021
- Real-time task scheduling for machine perception in intelligent cyber-physical systemsIEEE Transactions on Computers, 2021
- Contrastive self-supervised representation learning for sensing signals from the time-frequency perspectiveIn 2021 International Conference on Computer Communications and Networks (ICCCN), 2021
- Audio keyword reconstruction from on-device motion sensor signals via neural frequency unfoldingProceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, 2021
- Computational modeling of hierarchically polarized groups by structured matrix factorizationFrontiers in big Data, 2021
- Towards an accurate latency model for convolutional neural network layers on gpusIn MILCOM 2021-2021 IEEE Military Communications Conference (MILCOM), 2021
2020
- Multiscale online media simulation with socialcubeComputational and Mathematical Organization Theory, 2020
- Five challenges in cloud-enabled intelligence and controlACM Transactions on Internet Technology (TOIT), 2020
- Disentangling overlapping beliefs by structured matrix factorizationarXiv e-prints, pp. arXiv–2002, 2020
- DeepMV: Multi-view deep learning for device-free human activity recognitionProceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, 2020
- Giobalfusion: A global attentional deep learning framework for multisensor information fusionProceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, 2020
- Revisiting over-smoothing in deep GCNsarXiv preprint arXiv:2003.13663, 2020
- paper2repo: GitHub repository recommendation for academic papersIn Proceedings of The Web Conference 2020, 2020
- Truth discovery with multi-modal data in social sensingIEEE Transactions on Computers, 2020
- Handling missing sensors in topology-aware iot applications with gated graph neural networkProceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, 2020
- Dynamicvae: Decoupling reconstruction error and disentangled representation learningarXiv preprint arXiv:2009.06795, 2020
- Scheduling real-time deep learning services as imprecise computationsIn 2020 IEEE 26th international conference on embedded and real-time computing systems and applications (RTCSA), 2020
- Misinformation detection and adversarial attack cost analysis in directional social networksIn 2020 29th International Conference on Computer Communications and Networks (ICCCN), 2020
- On removing algorithmic priority inversion from mission-critical machine inference pipelinesIn 2020 IEEE Real-Time Systems Symposium (RTSS), 2020
- Controlvae: Controllable variational autoencoderIn International conference on machine learning, 2020
- Deep compressive offloading: Speeding up neural network inference by trading edge computation for network latencyIn Proceedings of the 18th conference on embedded networked sensor systems, 2020
- Hierarchical overlapping belief estimation by structured matrix factorizationIn 2020 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM), 2020
- Revisit oversmoothing in deep GCNsJournal of Environmental Sciences (China) English Ed, 2020
- AMVP: Adaptive CNN-based multitask video processing on mobile stream processing platformsIn 2020 IEEE/ACM Symposium on Edge Computing (SEC), 2020
2019
- STFNets: Learning sensing signals from the time-frequency perspective with short-time fourier neural networksIn The World Wide Web Conference, 2019
- A latent hawkes process model for event clustering and temporal dynamics learning with applications in githubIn 2019 IEEE 39th International Conference on Distributed Computing Systems (ICDCS), 2019
- Dependable machine intelligence at the tactical edgeIn Artificial Intelligence and Machine Learning for Multi-Domain Operations Applications, 2019
- DeepFusion: A deep learning framework for the fusion of heterogeneous sensory dataIn Proceedings of the Twentieth ACM International Symposium on Mobile Ad Hoc Networking and Computing, 2019
- Sadeepsense: Self-attention deep learning framework for heterogeneous on-device sensors in internet of things applicationsIn IEEE INFOCOm 2019-IEEE conference on computer communications, 2019
- Greenroute: a generalizable fuel-saving vehicular navigation serviceIn 2019 IEEE International Conference on Autonomic Computing (ICAC), 2019
- Simulation evaluation of fuel-saving systems in the city of ChicagoIn 2019 28th International Conference on Computer Communication and Networks (ICCCN), 2019
- Eugene: Towards deep intelligence as a serviceIn 2019 IEEE 39th International Conference on Distributed Computing Systems (ICDCS), 2019
- Stardust: A deep learning serving system in IoT: Demo abstractIn Proceedings of the 17th Conference on Embedded Networked Sensor Systems, 2019
- Unsupervised fact-finding with multi-modal data in social sensingIn 2019 22th International Conference on Information Fusion (FUSION), 2019
2018
- Rdeepsense: Reliable deep mobile computing models with uncertainty estimationsProceedings of the ACM on interactive, mobile, wearable and ubiquitous technologies, 2018
- Athena: Towards decision-centric anticipatory sensor information deliveryJournal of Sensor and Actuator Networks, 2018
- A spherical hidden Markov model for semantics-rich human mobility modelingIn Proceedings of the AAAI conference on artificial intelligence, 2018
- Codrive: Cooperative driving scheme for vehicles in urban signalized intersectionsIn 2018 ACM/IEEE 9th International Conference on Cyber-Physical Systems (ICCPS), 2018
- Deep learning for the internet of thingsComputer, 2018
- De facto diagnosis specialties: R ecognition and discoveryLearning Health Systems, 2018
- Qualitydeepsense: Quality-aware deep learning framework for internet of things applications with sensor-temporal attentionIn Proceedings of the 2nd International Workshop on Embedded and Mobile Deep Learning, 2018
- A constrained maximum likelihood estimator for unguided social sensingIn IEEE INFOCOM 2018-IEEE Conference on Computer Communications, 2018
- ApDeepsense: Deep learning uncertainty estimation without the pain for iot applicationsIn 2018 IEEE 38th International Conference on Distributed Computing Systems (ICDCS), 2018
- Towards environment independent device free human activity recognitionIn Proceedings of the 24th annual international conference on mobile computing and networking, 2018
- SenseGAN: Enabling deep learning for internet of things with a semi-supervised frameworkProceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, 2018
- Fastdeepiot: Towards understanding and optimizing neural network execution time on mobile and embedded devicesIn Proceedings of the 16th ACM Conference on Embedded Networked Sensor Systems, 2018
- A predictive self-configuring simulator for online mediaIn 2018 Winter Simulation Conference (WSC), 2018
2017
- Deepsense: A unified deep learning framework for time-series mobile sensing data processingIn Proceedings of the 26th international conference on world wide web, 2017
- Unveiling polarization in social networks: A matrix factorization approachIn IEEE INFOCOM 2017-IEEE Conference on Computer Communications, 2017
- Greendrive: A smartphone-based intelligent speed adaptation system with real-time traffic signal predictionIn Proceedings of the 8th International Conference on Cyber-Physical Systems, 2017
- DeepIoT: Compressing Deep Neural Network Structures for Sensing Systems with a Compressor-Critic FrameworkIn SenSys ’17 Proceedings of the 15th ACM Conference on Embedded Network Sensor Systems, 2017
- Unsupervised Fill-level Estimation for Smart Trash Removal Systems.In EWSN, 2017
- Optimizing source selection in social sensing in the presence of influence graphsIn 2017 IEEE 37th International Conference on Distributed Computing Systems (ICDCS), 2017
- Decision-driven execution: A distributed resource management paradigm for the age of iotIn 2017 IEEE 37th International Conference on Distributed Computing Systems (ICDCS), 2017
- VibeBin: A vibration-based waste bin level detection systemProceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, 2017
- On the improvement of classifying EEG recordings using neural networksIn 2017 IEEE International Conference on Big Data (Big Data), 2017
2016
- Recursive ground truth estimator for social data streamsIn 2016 15th ACM/IEEE International Conference on Information Processing in Sensor Networks (IPSN), 2016
- An experimental evaluation of datacenter workloads on low-power embedded micro serversProceedings of the VLDB Endowment 9, no. 9 (2016): 696-707., 2016
- On source dependency models for reliable social sensing: Algorithms and fundamental error boundsIn 2016 IEEE 36th international conference on distributed computing systems (ICDCS), 2016
- Impacts of social relationships and inhomogeneous node distribution on the network performanceIEEE Transactions on Wireless Communications, 2016
- Optimal capacity–delay tradeoff in MANETs with correlation of node mobilityIEEE Transactions on Vehicular Technology, 2016
2015
- Scalable social sensing of interdependent phenomenaIn Proceedings of the 14th international conference on information processing in sensor networks, 2015
- On exploiting logical dependencies for minimizing additive cost metrics in resource-limited crowdsensingIn 2015 International Conference on Distributed Computing in Sensor Systems, 2015
- Data acquisition for real-time decision-making under freshness constraintsIn 2015 IEEE Real-Time Systems Symposium, 2015
-
2014
- Delay-throughput tradeoff with correlated mobility of ad-hoc networksIn IEEE INFOCOM 2014-IEEE Conference on Computer Communications, 2014