{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,13]],"date-time":"2026-05-13T17:23:23Z","timestamp":1778693003809,"version":"3.51.4"},"reference-count":124,"publisher":"Association for Computing Machinery (ACM)","issue":"2","license":[{"start":{"date-parts":[[2025,1,21]],"date-time":"2025-01-21T00:00:00Z","timestamp":1737417600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"crossref","award":["62325201 and 62102009"],"award-info":[{"award-number":["62325201 and 62102009"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Softw. Eng. Methodol."],"published-print":{"date-parts":[[2025,2,28]]},"abstract":"<jats:p>\n            Web applications have increasingly adopted Deep Learning (DL) through\n            <jats:italic>in-browser inference<\/jats:italic>\n            , wherein DL inference performs directly within Web browsers. The actual performance of in-browser inference and its impacts on the Quality of Experience (\n            <jats:italic>QoE<\/jats:italic>\n            ) remain unexplored, and urgently require new QoE measurements beyond traditional ones, e.g., mainly focusing on page load time. To bridge this gap, we make the first comprehensive performance measurement of in-browser inference to date. Our approach proposes new metrics to measure in-browser inference: responsiveness, smoothness, and inference accuracy. Our extensive analysis involves 9 representative DL models across Web browsers of 50 popular PC devices and 20 mobile devices. The results reveal that in-browser inference exhibits a substantial latency gap, averaging 16.9 times slower on CPU and 4.9 times slower on GPU compared to native inference on PC devices. The gap on mobile CPU and mobile GPU is 15.8 times and 7.8 times, respectively. Furthermore, we identify contributing factors to such latency gap, including underutilized hardware instruction sets, inherent overhead in the runtime environment, resource contention within the browser, and inefficiencies in software libraries and GPU abstractions. Additionally, in-browser inference imposes significant memory demands, at times exceeding 334.6 times the size of the DL models themselves, partly attributable to suboptimal memory management. We also observe that in-browser inference leads to a significant 67.2% increase in the time it takes for GUI components to render within Web browsers, significantly affecting the overall user QoE of Web applications reliant on this technology.\n          <\/jats:p>","DOI":"10.1145\/3688843","type":"journal-article","created":{"date-parts":[[2024,8,14]],"date-time":"2024-08-14T15:16:35Z","timestamp":1723648595000},"page":"1-43","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":7,"title":["Anatomizing Deep Learning Inference in Web Browsers"],"prefix":"10.1145","volume":"34","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-1588-0293","authenticated-orcid":false,"given":"Qipeng","family":"Wang","sequence":"first","affiliation":[{"name":"Peking University, Beijing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4685-9633","authenticated-orcid":false,"given":"Shiqi","family":"Jiang","sequence":"additional","affiliation":[{"name":"Microsoft Research Asia, Beijing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4765-1893","authenticated-orcid":false,"given":"Zhenpeng","family":"Chen","sequence":"additional","affiliation":[{"name":"Nanyang Technological University, Singapore, Singapore"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-1866-1909","authenticated-orcid":false,"given":"Xu","family":"Cao","sequence":"additional","affiliation":[{"name":"Microsoft Research Asia, Beijing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1591-2526","authenticated-orcid":false,"given":"Yuanchun","family":"Li","sequence":"additional","affiliation":[{"name":"Institute for AI Industry Research (AIR), Tsinghua University, Beijing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-2703-1204","authenticated-orcid":false,"given":"Aoyu","family":"Li","sequence":"additional","affiliation":[{"name":"Nanyang Technological University, Singapore, Singapore"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7866-4075","authenticated-orcid":false,"given":"Yun","family":"Ma","sequence":"additional","affiliation":[{"name":"Peking University, Beijing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9107-013X","authenticated-orcid":false,"given":"Ting","family":"Cao","sequence":"additional","affiliation":[{"name":"Microsoft Research Asia, Beijing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7908-8484","authenticated-orcid":false,"given":"Xuanzhe","family":"Liu","sequence":"additional","affiliation":[{"name":"Peking University, Beijing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"320","published-online":{"date-parts":[[2025,1,21]]},"reference":[{"key":"e_1_3_2_2_2","unstructured":"ANGLE (software). 2024. Retrieved April 2024 from https:\/\/en.wikipedia.org\/wiki\/ANGLE_(software)"},{"key":"e_1_3_2_3_2","unstructured":"Selenium with Python. 2024. Retrieved April 2024 from https:\/\/selenium-python.readthedocs.io\/"},{"key":"e_1_3_2_4_2","unstructured":"Using the SavedModel Format. 2024. Retrieved April 2024 from https:\/\/www.tensorflow.org\/guide\/saved_model"},{"key":"e_1_3_2_5_2","unstructured":"WEBGL_debug_renderer_info Extension. 2024. Retrieved April 2024 from https:\/\/developer.mozilla.org\/en-US\/docs\/Web\/API\/WEBGL_debug_renderer_info"},{"key":"e_1_3_2_6_2","unstructured":"XNNPACK: High-Efficiency Floating-Point Neural Network Inference Operators for Mobile Server and Web. 2024. Retrieved April 2024 from https:\/\/github.com\/google\/XNNPACK"},{"key":"e_1_3_2_7_2","unstructured":"Grammar Checking. 2024. Retrieved July 2024 from https:\/\/www.microsoft.com\/en-us\/microsoft-365\/microsoft-editor\/grammar-checker"},{"key":"e_1_3_2_8_2","unstructured":"Teams Online. 2024. Retrieved July 2024 from https:\/\/teams.microsoft.com\/"},{"key":"e_1_3_2_9_2","unstructured":"Theremix. 2024. Retrieved July 2024 from https:\/\/theremin.app\/"},{"key":"e_1_3_2_10_2","unstructured":"AI Benchmark for Windows Linux and macOS. 2023. Retrieved September 2023 from https:\/\/ai-benchmark.com\/ranking_deeplearning_detailed.html"},{"key":"e_1_3_2_11_2","unstructured":"AI Face Test. 2023. Retrieved September 2023 from https:\/\/facetest1.netlify.app\/"},{"key":"e_1_3_2_12_2","unstructured":"Brain.js: GPU Accelerated Neural networks in JavaScript for Browsers and Node.js. 2023. Retrieved September 2023 from https:\/\/brain.js.org\/"},{"key":"e_1_3_2_13_2","unstructured":"Browser Market Share Worldwide. 2023. Retrieved September 2023 from https:\/\/gs.statcounter.com\/browser-market-share"},{"key":"e_1_3_2_14_2","unstructured":"Change Your Background in Microsoft Teams Meetings. 2023. Retrieved September 2023 from https:\/\/support.microsoft.com\/en-us\/office\/change-your-background-in-microsoft-teams-meetings-f77a2381-443a-499d-825e-509a140f4780"},{"key":"e_1_3_2_15_2","unstructured":"Core ML. 2023. Retrieved September 2023 from https:\/\/developer.apple.com\/documentation\/coreml"},{"key":"e_1_3_2_16_2","unstructured":"CPU Benchmarks \u2013 Over 1 000 000 CPUs benchmarked. 2023. Retrieved September 2023 from https:\/\/www.cpubenchmark.net\/"},{"key":"e_1_3_2_17_2","unstructured":"CUDA. 2023. Retrieved September 2023 from https:\/\/en.wikipedia.org\/wiki\/CUDA"},{"key":"e_1_3_2_18_2","unstructured":"Direct3D. 2023. Retrieved September 2023 from https:\/\/en.wikipedia.org\/wiki\/Direct3D"},{"key":"e_1_3_2_19_2","unstructured":"Face Age Test. 2023. Retrieved September 2023 from https:\/\/aige.netlify.app\/"},{"key":"e_1_3_2_20_2","unstructured":"FMA Instruction Set. 2023. Retrieved September 2023 from https:\/\/en.wikipedia.org\/wiki\/FMA_instruction_set"},{"key":"e_1_3_2_21_2","unstructured":"Getting Started Converting TensorFlow to ONNX. 2023. Retrieved September 2023 from https:\/\/onnxruntime.ai\/docs\/tutorials\/tf-get-started.html"},{"key":"e_1_3_2_22_2","unstructured":"How WebAssembly Threads Work. 2023. Retrieved September 2023 from https:\/\/web.dev\/webassembly-threads\/"},{"key":"e_1_3_2_23_2","unstructured":"Intel AVX-512 Instructions. 2023. Retrieved September 2023 from https:\/\/www.intel.com\/content\/www\/us\/en\/developer\/articles\/technical\/intel-avx-512-instructions.html"},{"key":"e_1_3_2_24_2","unstructured":"Intel AVX2 Instructions. 2023. Retrieved September 2023 from https:\/\/www.intel.com\/content\/www\/us\/en\/develop\/documentation\/cpp-compiler-developer-guide-and-reference\/top\/compiler-reference\/intrinsics\/intrinsics-for-avx2.html"},{"key":"e_1_3_2_25_2","unstructured":"Lighthouse. 2023. Retrieved September 2023 from https:\/\/developer.chrome.com\/docs\/lighthouse\/overview\/"},{"key":"e_1_3_2_26_2","unstructured":"Metal: Accelerating Graphics and Much More. 2023. Retrieved September 2023 from https:\/\/developer.apple.com\/metal\/"},{"key":"e_1_3_2_27_2","unstructured":"Model Conversion. 2023. Retrieved September 2023 from https:\/\/www.tensorflow.org\/js\/guide\/conversion"},{"key":"e_1_3_2_28_2","unstructured":"Module: tf.keras.applications. 2023. Retrieved September 2023 from https:\/\/www.tensorflow.org\/api_docs\/python\/tf\/keras\/applications"},{"key":"e_1_3_2_29_2","unstructured":"Navigation Timing API. 2023. Retrieved September 2023 from https:\/\/developer.mozilla.org\/en-US\/docs\/Web\/API\/Navigation_timing_API"},{"key":"e_1_3_2_30_2","unstructured":"ONNX. 2023. Retrieved September 2023 from https:\/\/onnx.ai\/"},{"key":"e_1_3_2_31_2","unstructured":"ONNX Runtime: Optimize and Accelerate Machine Learning Inferencing and Training. 2023. Retrieved September 2023 from https:\/\/onnxruntime.ai\/"},{"key":"e_1_3_2_32_2","unstructured":"ONNX Web Runtime. 2023. Retrieved September 2023 from https:\/\/onnxruntime.ai\/"},{"key":"e_1_3_2_33_2","unstructured":"OpenGL: The Industry\u2019s Foundation for High Performance Graphics. 2023. Retrieved September 2023 from https:\/\/www.opengl.org\/"},{"key":"e_1_3_2_34_2","unstructured":"Relaxed SIMD Proposal. 2023. Retrieved September 2023 from https:\/\/github.com\/WebAssembly\/relaxed-simd\/blob\/main\/proposals\/relaxed-simd\/Overview.md"},{"key":"e_1_3_2_35_2","unstructured":"Rocket Lake. 2023. Retrieved September 2023 from https:\/\/en.wikipedia.org\/wiki\/Rocket_Lake"},{"key":"e_1_3_2_36_2","unstructured":"Skylake (Microarchitecture). 2023. Retrieved September 2023 from https:\/\/en.wikipedia.org\/wiki\/Skylake_(microarchitecture)"},{"key":"e_1_3_2_37_2","unstructured":"Speed Index. 2023. Retrieved September 2023 from https:\/\/developer.chrome.com\/docs\/lighthouse\/performance\/speed-index\/"},{"key":"e_1_3_2_38_2","unstructured":"Speedometer. 2023. Retrieved September 2023 from https:\/\/browserbench.org\/Speedometer2.1\/"},{"key":"e_1_3_2_39_2","unstructured":"Technical City: Unbiased Hardware Comparisons. 2023. Retrieved September 2023 from https:\/\/technical.city\/"},{"key":"e_1_3_2_40_2","unstructured":"TensorFlow Hub. 2023. Retrieved September 2023 from https:\/\/tfhub.dev\/"},{"key":"e_1_3_2_41_2","unstructured":"TensorFlow Lite. 2023. Retrieved September 2023 from https:\/\/www.tensorflow.org\/lite"},{"key":"e_1_3_2_42_2","unstructured":"TensorFlow.js. 2023. Retrieved September 2023 from https:\/\/www.tensorflow.org\/js"},{"key":"e_1_3_2_43_2","unstructured":"Time to First Byte. 2023. Retrieved September 2023 from https:\/\/developer.mozilla.org\/en-US\/docs\/Glossary\/time_to_first_byte"},{"key":"e_1_3_2_44_2","unstructured":"Using Web Workers. 2023. Retrieved September 2023 from https:\/\/developer.mozilla.org\/en-US\/docs\/Web\/API\/Web_Workers_API\/Using_web_workers"},{"key":"e_1_3_2_45_2","unstructured":"V8. 2023. Retrieved September 2023 from https:\/\/v8.dev\/"},{"key":"e_1_3_2_46_2","unstructured":"Web LLM. 2023. Retrieved September 2023 from https:\/\/github.com\/mlc-ai\/web-llm"},{"key":"e_1_3_2_47_2","unstructured":"WebAssembly. 2023. Retrieved September 2023 from https:\/\/webassembly.org\/"},{"key":"e_1_3_2_48_2","unstructured":"WebAssembly 128-Bit Packed SIMD Extension. 2023. Retrieved September 2023 from https:\/\/github.com\/WebAssembly\/spec\/blob\/main\/proposals\/simd\/SIMD.md"},{"key":"e_1_3_2_49_2","unstructured":"WebDNN. 2023. Retrieved September 2023 from https:\/\/mil-tokyo.github.io\/webdnn\/"},{"key":"e_1_3_2_50_2","unstructured":"WebGL: 2D and 3D Graphics for the Web. 2023. Retrieved September 2023 from https:\/\/developer.mozilla.org\/en-US\/docs\/Web\/API\/WebGL_API"},{"key":"e_1_3_2_51_2","unstructured":"WebGL2RenderingContext. 2023. Retrieved September 2023 from https:\/\/developer.mozilla.org\/en-US\/docs\/Web\/API\/WebGL2RenderingContext"},{"key":"e_1_3_2_52_2","unstructured":"WebGL2RenderingContext: getSyncParameter() method. 2023. Retrieved September 2023 from https:\/\/developer.mozilla.org\/en-US\/docs\/Web\/API\/WebGL2RenderingContext\/getSyncParameter"},{"key":"e_1_3_2_53_2","unstructured":"WebGLShader. 2023. Retrieved September 2023 from https:\/\/developer.mozilla.org\/en-US\/docs\/Web\/API\/WebGLShader"},{"key":"e_1_3_2_54_2","unstructured":"WebGPU-W3C Working Draft. 2023. Retrieved September 2023 from https:\/\/www.w3.org\/TR\/webgpu\/"},{"key":"e_1_3_2_55_2","unstructured":"YouTube-BoundingBoxes Dataset. 2023. Retrieved September 2023 from https:\/\/research.google.com\/youtube-bb\/"},{"key":"e_1_3_2_56_2","unstructured":"YouTube Faces DB. 2023. Retrieved September 2023 from https:\/\/www.cs.tau.ac.il\/wolf\/ytfaces\/"},{"key":"e_1_3_2_57_2","first-page":"265","volume-title":"12th USENIX Symposium on Operating Systems Design and Implementation (OSDI \u201916)","author":"Abadi Mart\u00edn","year":"2016","unstructured":"Mart\u00edn Abadi, Paul Barham, Jianmin Chen, Zhifeng Chen, Andy Davis, Jeffrey Dean, Matthieu Devin, Sanjay Ghemawat, Geoffrey Irving, Michael Isard, Manjunath Kudlur, Josh Levenberg, Rajat Monga, Sherry Moore, Derek G. Murray, Benoit Steiner, Paul Tucker, Vijay Vasudevan, Pete Warden, Martin Wicke, Yuan Yu, and Xiaoqiang Zheng. 2016. TensorFlow: A system for large-scale machine learning. In 12th USENIX Symposium on Operating Systems Design and Implementation (OSDI \u201916), 265\u2013283."},{"key":"e_1_3_2_58_2","doi-asserted-by":"crossref","unstructured":"Amin Ahmad Noah Constant Yinfei Yang and Daniel Cer. 2019. ReQA: An evaluation for end-to-end answer retrieval models. arXiv:1907.04780. Retrieved from https:\/\/doi.org\/10.48550\/arXiv.1907.04780","DOI":"10.18653\/v1\/D19-5819"},{"key":"e_1_3_2_59_2","doi-asserted-by":"publisher","DOI":"10.1109\/RTSS.2018.00017"},{"key":"e_1_3_2_60_2","doi-asserted-by":"publisher","DOI":"10.1145\/3027947.3027949"},{"key":"e_1_3_2_61_2","first-page":"374","volume-title":"Proceedings of Machine Learning and Systems","volume":"1","author":"Bonawitz Keith","year":"2019","unstructured":"Keith Bonawitz, Hubert Eichner, Wolfgang Grieskamp, Dzmitry Huba, Alex Ingerman, Vladimir Ivanov, Chloe Kiddon, Jakub Kone\u010dn\u1ef3, Stefano Mazzocchi, Brendan McMahan, Timon Van Overveldt, David Petrou, Daniel Ramage, and Jason Roselander. 2019. Towards federated learning at scale: System design. Proceedings of Machine Learning and Systems 1 (2019), 374\u2013388."},{"key":"e_1_3_2_62_2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.01172"},{"key":"e_1_3_2_63_2","first-page":"406","volume-title":"Proceedings of Machine Learning and Systems","volume":"1","author":"Canel Christopher","year":"2019","unstructured":"Christopher Canel, Thomas Kim, Giulio Zhou, Conglong Li, Hyeontaek Lim, David G Andersen, Michael Kaminsky, and Subramanya Dulloor. 2019. Scaling video analytics on constrained edge nodes. Proceedings of Machine Learning and Systems 1 (2019), 406\u2013417."},{"key":"e_1_3_2_64_2","doi-asserted-by":"publisher","DOI":"10.1145\/3540250.3549123"},{"key":"e_1_3_2_65_2","first-page":"1","volume-title":"37th IEEE\/ACM International Conference on Automated Software Engineering","author":"Chen Simin","year":"2022","unstructured":"Simin Chen, Mirazul Haque, Cong Liu, and Wei Yang. 2022. DeepPerform: An efficient approach for performance testing of resource-constrained neural networks. In 37th IEEE\/ACM International Conference on Automated Software Engineering, 1\u201313."},{"key":"e_1_3_2_66_2","first-page":"578","volume-title":"13th USENIX Symposium on Operating Systems Design and Implementation (OSDI \u201918)","author":"Chen Tianqi","year":"2018","unstructured":"Tianqi Chen, Thierry Moreau, Ziheng Jiang, Lianmin Zheng, Eddie Yan, Haichen Shen, Meghan Cowan, Leyuan Wang, Yuwei Hu, Luis Ceze, Carlos Guestrin, and Arvind Krishnamurthy. 2018. TVM: An automated end-to-end optimizing compiler for deep learning. In 13th USENIX Symposium on Operating Systems Design and Implementation (OSDI \u201918), 578\u2013594."},{"key":"e_1_3_2_67_2","doi-asserted-by":"publisher","DOI":"10.1145\/2809695.2809711"},{"key":"e_1_3_2_68_2","doi-asserted-by":"publisher","DOI":"10.1145\/3368089.3409759"},{"key":"e_1_3_2_69_2","doi-asserted-by":"crossref","first-page":"674","DOI":"10.1109\/ICSE43902.2021.00068","volume-title":"2021 IEEE\/ACM 43rd International Conference on Software Engineering (ICSE)","author":"Chen Zhenpeng","year":"2021","unstructured":"Zhenpeng Chen, Huihan Yao, Yiling Lou, Yanbin Cao, Yuanqiang Liu, Haoyu Wang, and Xuanzhe Liu. 2021. An empirical study on deployment faults of deep learning based mobile applications. In 2021 IEEE\/ACM 43rd International Conference on Software Engineering (ICSE). IEEE, 674\u2013685."},{"key":"e_1_3_2_70_2","doi-asserted-by":"crossref","first-page":"1584","DOI":"10.1145\/3377811.3380404","volume-title":"ACM\/IEEE 42nd International Conference on Software Engineering","author":"Cummaudo Alex","year":"2020","unstructured":"Alex Cummaudo, Rajesh Vasa, Scott Barnett, John Grundy, and Mohamed Abdelrazek. 2020. Interpreting cloud computer vision pain-points: A mining study of stack overflow. In ACM\/IEEE 42nd International Conference on Software Engineering, 1584\u20131596."},{"key":"e_1_3_2_71_2","unstructured":"Jacob Devlin Ming-Wei Chang Kenton Lee and Kristina Toutanova. 2018. Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv:1810.04805. Retrieved from https:\/\/doi.org\/10.48550\/arXiv.1810.04805"},{"key":"e_1_3_2_72_2","doi-asserted-by":"crossref","first-page":"2499","DOI":"10.1109\/ICPADS60453.2023.00333","volume-title":"2023 IEEE 29th International Conference on Parallel and Distributed Systems (ICPADS)","author":"Dong Bing","year":"2023","unstructured":"Bing Dong, Tianen Liu, Borui Li, Xiaolei Zhou, Shuai Wang, and Zhao-Dong Xu. 2023. WebInf: Accelerating WebGPU-based in-browser DNN inference via adaptive model partitioning. In 2023 IEEE 29th International Conference on Parallel and Distributed Systems (ICPADS). IEEE, 2499\u20132506."},{"key":"e_1_3_2_73_2","doi-asserted-by":"publisher","DOI":"10.1145\/3338906.3338954"},{"key":"e_1_3_2_74_2","first-page":"5374","volume-title":"IEEE\/CVF Conference on Computer Vision and Pattern Recognition","author":"Fan Heng","year":"2019","unstructured":"Heng Fan, Liting Lin, Fan Yang, Peng Chu, Ge Deng, Sijia Yu, Hexin Bai, Yong Xu, Chunyuan Liao, and Haibin Ling. 2019. Lasot: A high-quality benchmark for large-scale single object tracking. In IEEE\/CVF Conference on Computer Vision and Pattern Recognition, 5374\u20135383."},{"key":"e_1_3_2_75_2","first-page":"1342","volume-title":"28th ACM Joint Meeting on European Software Engineering Conference and Symposium on the Foundations of Software Engineering","author":"Gao Yanjie","year":"2020","unstructured":"Yanjie Gao, Yu Liu, Hongyu Zhang, Zhengxian Li, Yonghao Zhu, Haoxiang Lin, and Mao Yang. 2020. Estimating GPU memory consumption of deep learning models. In 28th ACM Joint Meeting on European Software Engineering Conference and Symposium on the Foundations of Software Engineering, 1342\u20131352."},{"key":"e_1_3_2_76_2","doi-asserted-by":"publisher","DOI":"10.1109\/ASE.2019.00080"},{"key":"e_1_3_2_77_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICSE.2019.00113"},{"key":"e_1_3_2_78_2","doi-asserted-by":"publisher","DOI":"10.1145\/3062341.3062363"},{"key":"e_1_3_2_79_2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"e_1_3_2_80_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00140"},{"issue":"4","key":"e_1_3_2_81_2","doi-asserted-by":"crossref","first-page":"1950","DOI":"10.1109\/TMC.2021.3122279","article-title":"An integrated cloud-edge-device adaptive deep learning service for cross-platform web","volume":"22","author":"Huang Yakun","year":"2021","unstructured":"Yakun Huang, Xiuquan Qiao, Jian Tang, Pei Ren, Ling Liu, Calton Pu, and Jun-Liang Chen. 2021. An integrated cloud-edge-device adaptive deep learning service for cross-platform web. IEEE Transactions on Mobile Computing 22, 4 (2021), 1950\u20131967.","journal-title":"IEEE Transactions on Mobile Computing"},{"key":"e_1_3_2_82_2","first-page":"107","volume-title":"2019 USENIX Annual Technical Conference (USENIX ATC \u201919)","author":"Jangda Abhinav","year":"2019","unstructured":"Abhinav Jangda, Bobby Powers, Emery D Berger, and Arjun Guha. 2019. Not so fast: Analyzing the performance of WebAssembly vs. native code. In 2019 USENIX Annual Technical Conference (USENIX ATC \u201919), 107\u2013120."},{"key":"e_1_3_2_83_2","doi-asserted-by":"publisher","DOI":"10.1145\/3643832.3661892"},{"key":"e_1_3_2_84_2","doi-asserted-by":"crossref","unstructured":"Daniel Kang John Emmons Firas Abuzaid Peter Bailis and Matei Zaharia. 2017. NoScope: Optimizing neural network queries over video at scale. arXiv:1703.02529. Retrieved from https:\/\/doi.org\/10.48550\/arXiv.1703.02529","DOI":"10.14778\/3137628.3137664"},{"key":"e_1_3_2_85_2","unstructured":"Tero Karras Timo Aila Samuli Laine and Jaakko Lehtinen. 2017. Progressive growing of GANs for improved quality stability and variation. arXiv:1710.10196. Retrieved from https:\/\/doi.org\/10.48550\/arXiv.1710.10196"},{"key":"e_1_3_2_86_2","first-page":"545","volume-title":"14th USENIX Symposium on Networked Systems Design and Implementation (NSDI \u201917)","author":"Kelton Conor","year":"2017","unstructured":"Conor Kelton, Jihoon Ryoo, Aruna Balasubramanian, and Samir R. Das. 2017. Improving user perceived page load times using gaze. In 14th USENIX Symposium on Networked Systems Design and Implementation (NSDI \u201917), 545\u2013559."},{"key":"e_1_3_2_87_2","first-page":"359","volume-title":"Annual Conference of the ACM Special Interest Group on Data Communication on the Applications, Technologies, Architectures, and Protocols for Computer Communication","author":"Li Yuanqi","year":"2020","unstructured":"Yuanqi Li, Arthi Padmanabhan, Pengzhan Zhao, Yufei Wang, Guoqing Harry Xu, and Ravi Netravali. 2020. Reducto: On-camera filtering for resource-efficient real-time video analytics. In Annual Conference of the ACM Special Interest Group on Data Communication on the Applications, Technologies, Architectures, and Protocols for Computer Communication, 359\u2013376."},{"key":"e_1_3_2_88_2","first-page":"69","volume-title":"Companion Proceedings of the the Web Conference 2018","author":"Ling Chen","year":"2018","unstructured":"Chen Ling, Lei Wang, Jun Lang, Qiufen Xia, Guoxuan Chang, Kun Wang, and Peng Zhao. 2018. LinCa: A page loading time optimization approach for users subject to Internet access restriction. In Companion Proceedings of the the Web Conference 2018, 69\u201370."},{"key":"e_1_3_2_89_2","doi-asserted-by":"publisher","DOI":"10.1145\/3485447.3511981"},{"key":"e_1_3_2_90_2","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46448-0_2"},{"key":"e_1_3_2_91_2","first-page":"1","volume-title":"Proceedings of the ACM on Measurement and Analysis of Computing Systems","volume":"6","author":"Liu Wei","year":"2022","unstructured":"Wei Liu, Xinlei Yang, Hao Lin, Zhenhua Li, and Feng Qian. 2022. Fusing speed index during Web page loading. Proceedings of the ACM on Measurement and Analysis of Computing Systems 6, 1 (2022), 1\u201323."},{"key":"e_1_3_2_92_2","first-page":"6950","volume-title":"IEEE\/CVF International Conference on Computer Vision","author":"Liu Xin","year":"2023","unstructured":"Xin Liu, Fatemeh Karimi Nejadasl, Jan C. van Gemert, Olaf Booij, and Silvia L. Pintea. 2023. Objects do not disappear: Video object detection by single-frame object location anticipation. In IEEE\/CVF International Conference on Computer Vision, 6950\u20136961."},{"key":"e_1_3_2_93_2","doi-asserted-by":"publisher","DOI":"10.1145\/3308558.3313639"},{"key":"e_1_3_2_94_2","first-page":"461","volume-title":"15th USENIX Symposium on Operating Systems Design and Implementation OSDI \u201921","author":"Mardani Shaghayegh","year":"2021","unstructured":"Shaghayegh Mardani, Ayush Goel, Ronny Ko, Harsha V Madhyastha, and Ravi Netravali. 2021. Horcrux: Automatic JavaScript parallelism for resource-efficient web computation. In 15th USENIX Symposium on Operating Systems Design and Implementation OSDI \u201921, 461\u2013477."},{"key":"e_1_3_2_95_2","doi-asserted-by":"crossref","first-page":"940","DOI":"10.1145\/2971648.2971731","volume-title":"2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing","author":"Mittal Gaurav","year":"2016","unstructured":"Gaurav Mittal, Kaushal B. Yagnik, Mohit Garg, and Narayanan C. Krishnan. 2016. Spotgarbage: Smartphone app to detect garbage using deep learning. In 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing, 940\u2013945."},{"key":"e_1_3_2_96_2","unstructured":"Pavlo Molchanov Stephen Tyree Tero Karras Timo Aila and Jan Kautz. 2016. Pruning convolutional neural networks for resource efficient inference. arXiv:1611.06440. Retrieved from https:\/\/doi.org\/10.48550\/arXiv.1611.06440"},{"key":"e_1_3_2_97_2","first-page":"1913","volume-title":"IEEE International Conference on Computer Vision","author":"Pfister Tomas","year":"2015","unstructured":"Tomas Pfister, James Charles, and Andrew Zisserman. 2015. Flowing convnets for human pose estimation in videos. In IEEE International Conference on Computer Vision, 1913\u20131921."},{"key":"e_1_3_2_98_2","doi-asserted-by":"publisher","DOI":"10.1145\/3324884.3416545"},{"key":"e_1_3_2_99_2","doi-asserted-by":"publisher","DOI":"10.1145\/3431832.3431837"},{"key":"e_1_3_2_100_2","first-page":"1","volume-title":"37th IEEE\/ACM International Conference on Automated Software Engineering","author":"Quan Lili","year":"2022","unstructured":"Lili Quan, Qianyu Guo, Xiaofei Xie, Sen Chen, Xiaohong Li, and Yang Liu. 2022. Towards understanding the faults of JavaScript-based deep learning systems. In 37th IEEE\/ACM International Conference on Automated Software Engineering, 1\u201313."},{"key":"e_1_3_2_101_2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.91"},{"key":"e_1_3_2_102_2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00474"},{"key":"e_1_3_2_103_2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2015.7298682"},{"key":"e_1_3_2_104_2","unstructured":"Karen Simonyan and Andrew Zisserman. 2014. Very deep convolutional networks for large-scale image recognition. arXiv:1409.1556. Retrieved from https:\/\/doi.org\/10.48550\/arXiv.1409.1556"},{"key":"e_1_3_2_105_2","unstructured":"Zhiqing Sun Hongkun Yu Xiaodan Song Renjie Liu Yiming Yang and Denny Zhou. 2020. MobileBert: A compact task-agnostic Bert for resource-limited devices. arXiv:2004.02984. Retrieved from https:\/\/doi.org\/10.48550\/arXiv.2004.02984"},{"key":"e_1_3_2_106_2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01079"},{"key":"e_1_3_2_107_2","doi-asserted-by":"publisher","DOI":"10.1145\/3570361.3613285"},{"key":"e_1_3_2_108_2","doi-asserted-by":"publisher","DOI":"10.1145\/3361242.3361249"},{"key":"e_1_3_2_109_2","doi-asserted-by":"publisher","DOI":"10.1145\/3498361.3538763"},{"key":"e_1_3_2_110_2","doi-asserted-by":"publisher","DOI":"10.1109\/TSC.2014.2387751"},{"key":"e_1_3_2_111_2","first-page":"2717","volume-title":"IEEE\/CVF Conference on Computer Vision and Pattern Recognition","author":"Wadhwani Krishna","year":"2022","unstructured":"Krishna Wadhwani and Tamaki Kojima. 2022. SqueezeNeRF: Further factorized FastNeRF for memory-efficient inference. In IEEE\/CVF Conference on Computer Vision and Pattern Recognition, 2717\u20132725."},{"key":"e_1_3_2_112_2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00552"},{"key":"e_1_3_2_113_2","doi-asserted-by":"publisher","DOI":"10.1145\/3447993.3448625"},{"key":"e_1_3_2_114_2","doi-asserted-by":"publisher","DOI":"10.1145\/3498361.3538928"},{"key":"e_1_3_2_115_2","volume-title":"European Conference on Computer Vision (ECCV) Workshops","author":"Wang Xintao","year":"2018","unstructured":"Xintao Wang, Ke Yu, Shixiang Wu, Jinjin Gu, Yihao Liu, Chao Dong, Yu Qiao, and Chen Change Loy. 2018. ESRGAN: Enhanced super-resolution generative adversarial networks. In European Conference on Computer Vision (ECCV) Workshops."},{"key":"e_1_3_2_116_2","first-page":"473","volume-title":"10th USENIX Symposium on Networked Systems Design and Implementation (NSDI \u201913)","author":"Wang Xiao Sophia","year":"2013","unstructured":"Xiao Sophia Wang, Aruna Balasubramanian, Arvind Krishnamurthy, and David Wetherall. 2013. Demystifying page load performance with WProf. In 10th USENIX Symposium on Networked Systems Design and Implementation (NSDI \u201913), 473\u2013485."},{"key":"e_1_3_2_117_2","first-page":"1","volume-title":"Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies","volume":"2","author":"Xu Mengwei","year":"2018","unstructured":"Mengwei Xu, Feng Qian, Qiaozhu Mei, Kang Huang, and Xuanzhe Liu. 2018. DeepType: On-device deep learning for input personalization service with minimal privacy concern. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies 2, 4 (2018), 1\u201326."},{"key":"e_1_3_2_118_2","doi-asserted-by":"publisher","DOI":"10.1145\/3386901.3388948"},{"key":"e_1_3_2_119_2","doi-asserted-by":"publisher","DOI":"10.1145\/3543507.3583323"},{"key":"e_1_3_2_120_2","doi-asserted-by":"publisher","DOI":"10.1145\/3505243"},{"key":"e_1_3_2_121_2","doi-asserted-by":"publisher","DOI":"10.1145\/3308558.3313575"},{"key":"e_1_3_2_122_2","doi-asserted-by":"publisher","DOI":"10.1145\/3485447.3512148"},{"key":"e_1_3_2_123_2","doi-asserted-by":"publisher","DOI":"10.1109\/ISSRE.2019.00020"},{"key":"e_1_3_2_124_2","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA56546.2023.10071077"},{"key":"e_1_3_2_125_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.244"}],"container-title":["ACM Transactions on Software Engineering and Methodology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3688843","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3688843","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,19]],"date-time":"2025-06-19T00:04:10Z","timestamp":1750291450000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3688843"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,1,21]]},"references-count":124,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2025,2,28]]}},"alternative-id":["10.1145\/3688843"],"URL":"https:\/\/doi.org\/10.1145\/3688843","relation":{},"ISSN":["1049-331X","1557-7392"],"issn-type":[{"value":"1049-331X","type":"print"},{"value":"1557-7392","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,1,21]]},"assertion":[{"value":"2024-02-20","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2024-07-22","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-01-21","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}