{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,9]],"date-time":"2026-07-09T22:32:47Z","timestamp":1783636367346,"version":"3.55.0"},"reference-count":39,"publisher":"Association for Computing Machinery (ACM)","issue":"3","license":[{"start":{"date-parts":[[2026,5,13]],"date-time":"2026-05-13T00:00:00Z","timestamp":1778630400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/legalcode"}],"funder":[{"name":"National Science Foundation","award":["2513431"],"award-info":[{"award-number":["2513431"]}]},{"name":"U.S. Department of Energy (DOE)\u2019s Advanced Scientific Computing Research (ASCR) program"},{"name":"Office of Research Computing at George Mason University"},{"name":"National Science Foundation","award":["2018631"],"award-info":[{"award-number":["2018631"]}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Embed. Comput. Syst."],"published-print":{"date-parts":[[2026,5,31]]},"abstract":"<jats:p>\n                    Full-wave inversion (FWI), as a fundamental scientific approach to deducing unknown or unobservable subsurface properties, holds significant value in geophysics applications. Traditional FWI methods rely on physics-driven approaches that demand substantial computational resources. Recently, with the breakthroughs in machine learning (ML) and the prevalence of AI for science, data-driven approaches have been applied to FWI, showing promising results. However, as these applications often necessitate deployment in diverse regions with remote and extreme environments, localization of ML models on edge devices becomes imperative. A promising approach involves leveraging Generative AI models and governing wave equations to generate paired training data, including geophysical measurements (i.e., seismic waveform) as data and corresponding velocity maps as labels for model fine-tuning. However, the limited resources on edge devices pose significant challenges to achieving high software efficiency and low latency. In this article, we present a toolkit, namely DiGiT, a\n                    <jats:underline>di<\/jats:underline>\n                    ffusion-based modular\n                    <jats:underline>g<\/jats:underline>\n                    eophys\n                    <jats:underline>i<\/jats:underline>\n                    cal\n                    <jats:underline>t<\/jats:underline>\n                    oolkit platform. One key component is a library of decomposed modules from the widely used geophysical designs. Benefiting from the flexibility of combining modules, we composite a toolkit for the generation of on-device diffusion-based paired geophysical training data. The toolkit includes a 1-in-2-out network structure and diffusion model distillation, both of which can significantly reduce the computational time. Experiments on the OpenFWI dataset show that the DiGiT toolkit can generate paired seismic waveform and velocity map in seconds, which is over 100\u00d7 speedup compared with the sequential execution of the diffusion model and the wave equation-based forward modeling.\n                  <\/jats:p>","DOI":"10.1145\/3779425","type":"journal-article","created":{"date-parts":[[2025,12,5]],"date-time":"2025-12-05T22:29:24Z","timestamp":1764973764000},"page":"1-26","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["DiGiT: A Diffusion-based Modular Geophysical Toolkit for On-device Multi-modal Data Generation"],"prefix":"10.1145","volume":"25","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-4848-0248","authenticated-orcid":false,"given":"Junhuan","family":"Yang","sequence":"first","affiliation":[{"name":"George Mason University","place":["Fairfax, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-1116-4819","authenticated-orcid":false,"given":"Yuzhou","family":"Zhang","sequence":"additional","affiliation":[{"name":"Northeastern University","place":["Seattle, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1411-3554","authenticated-orcid":false,"given":"Yi","family":"Sheng","sequence":"additional","affiliation":[{"name":"George Mason University","place":["Fairfax, United States"]},{"name":"University of South Florida","place":["Fairfax, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7337-6760","authenticated-orcid":false,"given":"Youzuo","family":"Lin","sequence":"additional","affiliation":[{"name":"The University of North Carolina at Chapel Hill","place":["Chapel Hill, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9004-487X","authenticated-orcid":false,"given":"Weiwen","family":"Jiang","sequence":"additional","affiliation":[{"name":"George Mason University","place":["Fairfax, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0646-440X","authenticated-orcid":false,"given":"Lei","family":"Yang","sequence":"additional","affiliation":[{"name":"George Mason University","place":["Fairfax, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"320","published-online":{"date-parts":[[2026,5,13]]},"reference":[{"key":"e_1_3_1_2_2","doi-asserted-by":"publisher","DOI":"10.3390\/forecast3010012"},{"key":"e_1_3_1_3_2","unstructured":"Changyou Chen Han Ding Bunyamin Sisman Yi Xu Ouye Xie Benjamin Z. Yao Son Dinh Tran and Belinda Zeng. 2024. Diffusion models for multi-task generative modeling. arXiv:2407.17571. Retrieved from https:\/\/arxiv.org\/abs\/2407.17571"},{"key":"e_1_3_1_4_2","volume-title":"Proceedings of the Advances in Neural Information Processing Systems 35 (NeurIPS 2022) Datasets and Benchmarks Track","author":"Deng Chengyuan","year":"2022","unstructured":"Chengyuan Deng, Yinan Feng, Shihang Feng, Peng Jin, Xitong Zhang, Qili Zeng, and Youzuo Lin. 2022. OpenFWI: Benchmark seismic datasets for machine learning-based full waveform inversion. In Proceedings of the Advances in Neural Information Processing Systems 35 (NeurIPS 2022) Datasets and Benchmarks Track. Retrieved from https:\/\/openfwi-lanl.github.io\/"},{"key":"e_1_3_1_5_2","doi-asserted-by":"publisher","DOI":"10.1109\/PerCom59722.2024.10494442"},{"key":"e_1_3_1_6_2","first-page":"1","article-title":"Multiscale data-driven seismic full-waveform inversion with field data study","author":"Feng Shihang","year":"2021","unstructured":"Shihang Feng, Youzuo Lin, and Brendt Wohlberg. 2021. Multiscale data-driven seismic full-waveform inversion with field data study. IEEE Transactions on Geoscience and Remote Sensing 60 (2021), 1\u201314.","journal-title":"IEEE Transactions on Geoscience and Remote Sensing"},{"key":"e_1_3_1_7_2","volume-title":"Proceedings of the Advances in Neural Information Processing Systems 35 (NeurIPS 2022) Datasets and Benchmarks Track","author":"Feng Shihang","year":"2023","unstructured":"Shihang Feng, Hanchen Wang, Chengyuan Deng, Yinan Feng, Yanhua Liu, Min Zhu, Peng Jin, Yinpeng Chen, and Youzuo Lin. 2023. \\({E^{FWI}}\\) : Multi-parameter benchmark datasets for elastic full waveform inversion of geophysical properties. In Proceedings of the Advances in Neural Information Processing Systems 35 (NeurIPS 2022) Datasets and Benchmarks Track. Retrieved from https:\/\/efwi-lanl.github.io\/"},{"key":"e_1_3_1_8_2","first-page":"6434","volume-title":"Proceedings of the International Conference on Machine Learning","author":"Feng Yinan","year":"2022","unstructured":"Yinan Feng, Yinpeng Chen, Shihang Feng, Peng Jin, Zicheng Liu, and Youzuo Lin. 2022. An intriguing property of geophysics inversion. In Proceedings of the International Conference on Machine Learning. PMLR, 6434\u20136446."},{"key":"e_1_3_1_9_2","first-page":"6840","article-title":"Denoising diffusion probabilistic models","volume":"33","author":"Ho Jonathan","year":"2020","unstructured":"Jonathan Ho, Ajay Jain, and Pieter Abbeel. 2020. Denoising diffusion probabilistic models. Advances In Neural Information Processing Systems 33 (2020), 6840\u20136851.","journal-title":"Advances In Neural Information Processing Systems"},{"key":"e_1_3_1_10_2","doi-asserted-by":"publisher","DOI":"10.1038\/s42256-023-00659-9"},{"key":"e_1_3_1_11_2","volume-title":"Proceedings of the 10th International Conference on Learning Representations (ICLR)","author":"Jin Peng","year":"2022","unstructured":"Peng Jin, Xitong Zhang, Yinpeng Chen, Sharon Huang, Zicheng Liu, and Youzuo Lin. 2022. Unsupervised learning of full-waveform inversion: Connecting CNN and partial differential equation in a loop. In Proceedings of the 10th International Conference on Learning Representations (ICLR)."},{"key":"e_1_3_1_12_2","unstructured":"Peng Jin Xitong Zhang Yinpeng Chen Sharon Xiaolei Huang Zicheng Liu and Youzuo Lin. 2021. Unsupervised learning of full-waveform inversion: Connecting CNN and partial differential equation in a loop. arXiv:2110.07584. Retrieved from https:\/\/arxiv.org\/abs\/2110.07584"},{"key":"e_1_3_1_13_2","unstructured":"Diederik P. Kingma and Max Welling. 2013. Auto-encoding variational bayes. arXiv:1312.6114. Retrieved from https:\/\/arxiv.org\/abs\/1312.6114"},{"key":"e_1_3_1_14_2","first-page":"489","article-title":"Acoustic- and elastic-waveform inversion using a modified Total-Variation regularization scheme","volume":"200","author":"Lin Youzuo","year":"2015","unstructured":"Youzuo Lin and Lianjie Huang. 2015. Acoustic- and elastic-waveform inversion using a modified Total-Variation regularization scheme. Geophysical Journal International 200, 1 (2015), 489\u2013502.","journal-title":"Geophysical Journal International"},{"key":"e_1_3_1_15_2","first-page":"2125","article-title":"Quantifying subsurface geophysical properties changes using double-difference seismic-waveform inversion with a modified total-variation regularization scheme","volume":"203","author":"Lin Youzuo","year":"2016","unstructured":"Youzuo Lin and Lianjie Huang. 2016. Quantifying subsurface geophysical properties changes using double-difference seismic-waveform inversion with a modified total-variation regularization scheme. Geophysical Journal International 203, 3 (2016), 2125\u20132149.","journal-title":"Geophysical Journal International"},{"key":"e_1_3_1_16_2","doi-asserted-by":"publisher","DOI":"10.1109\/MSP.2022.3217658"},{"key":"e_1_3_1_17_2","article-title":"Seismic waveform inversion capability on resource-constrained edge devices","volume":"8","author":"Manu Daniel","year":"2022","unstructured":"Daniel Manu, Petro Tshakwanda, Youzuo Lin, Weiwen Jiang, and Lei Yang. 2022. Seismic waveform inversion capability on resource-constrained edge devices. Journal of Imaging 8, 12 (2022), 312.","journal-title":"Journal of Imaging"},{"key":"e_1_3_1_18_2","unstructured":"Robin Rombach Andreas Blattmann Dominik Lorenz Patrick Esser and Bj\u00f6rn Ommer. 2021. High-Resolution image synthesis with latent diffusion models. arxiv:2112.10752 [cs.CV]. Retrieved from https:\/\/arxiv.org\/abs\/2112.10752"},{"key":"e_1_3_1_19_2","unstructured":"Tim Salimans and Jonathan Ho. 2022. Progressive distillation for fast sampling of diffusion models. arXiv:2202.00512. Retrieved from https:\/\/arxiv.org\/abs\/2202.00512"},{"key":"e_1_3_1_20_2","doi-asserted-by":"publisher","DOI":"10.1145\/3489517.3530427"},{"key":"e_1_3_1_21_2","first-page":"2256","volume-title":"Proceedings of the International Conference on Machine Learning","author":"Sohl-Dickstein Jascha","year":"2015","unstructured":"Jascha Sohl-Dickstein, Eric Weiss, Niru Maheswaranathan, and Surya Ganguli. 2015. Deep unsupervised learning using nonequilibrium thermodynamics. In Proceedings of the International Conference on Machine Learning. PMLR, 2256\u20132265."},{"key":"e_1_3_1_22_2","unstructured":"Jiaming Song Chenlin Meng and Stefano Ermon. 2020. Denoising diffusion implicit models. arXiv:2010.02502. Retrieved from https:\/\/arxiv.org\/abs\/2010.02502"},{"key":"e_1_3_1_23_2","article-title":"Generative modeling by estimating gradients of the data distribution","volume":"32","author":"Song Yang","year":"2019","unstructured":"Yang Song and Stefano Ermon. 2019. Generative modeling by estimating gradients of the data distribution. Advances in Neural Information Processing Systems 32 (2019).","journal-title":"Advances in Neural Information Processing Systems"},{"key":"e_1_3_1_24_2","doi-asserted-by":"publisher","DOI":"10.1190\/geo2020-0312.1"},{"key":"e_1_3_1_25_2","doi-asserted-by":"publisher","DOI":"10.1190\/geo2019-0138.1"},{"key":"e_1_3_1_26_2","doi-asserted-by":"publisher","DOI":"10.1190\/1.3238367"},{"key":"e_1_3_1_27_2","doi-asserted-by":"publisher","DOI":"10.1109\/TGRS.2023.3335484"},{"key":"e_1_3_1_28_2","unstructured":"Hanchen Wang Yinpeng Chen Tariq Alkhalifah Ting Chen Youzuo Lin and David Alumbaugh. 2023. DeFault: Deep-learning-based fault delineation. arXiv:2311.04361. Retrieved from https:\/\/arxiv.org\/abs\/2311.04361"},{"key":"e_1_3_1_29_2","doi-asserted-by":"crossref","unstructured":"Zhepeng Wang Isaacshubhanand Putla Weiwen Jiang and Youzuo Lin. 2023. Edge-inversionnet: Enabling efficient inference of inversionnet on edge devices. arXiv:2310.09667. Retrieved from https:\/\/arxiv.org\/abs\/2310.09667","DOI":"10.1190\/image2023-3907875.1"},{"key":"e_1_3_1_30_2","doi-asserted-by":"publisher","DOI":"10.1190\/image2023-3907875.1"},{"key":"e_1_3_1_31_2","doi-asserted-by":"publisher","DOI":"10.1109\/TCI.2019.2956866"},{"key":"e_1_3_1_32_2","doi-asserted-by":"publisher","DOI":"10.1190\/geo2018-0249.1"},{"key":"e_1_3_1_33_2","doi-asserted-by":"publisher","DOI":"10.1109\/DAC56929.2023.10247959"},{"key":"e_1_3_1_34_2","doi-asserted-by":"publisher","DOI":"10.1145\/3649329.3657344"},{"key":"e_1_3_1_35_2","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v39i20.35504"},{"key":"e_1_3_1_36_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijggc.2019.102803"},{"key":"e_1_3_1_37_2","doi-asserted-by":"publisher","DOI":"10.1109\/TGRS.2022.3144636"},{"key":"e_1_3_1_38_2","doi-asserted-by":"publisher","DOI":"10.1109\/TGRS.2021.3135354"},{"key":"e_1_3_1_39_2","doi-asserted-by":"publisher","DOI":"10.1109\/TGRS.2020.2977635"},{"key":"e_1_3_1_40_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijggc.2019.102790"}],"container-title":["ACM Transactions on Embedded Computing Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3779425","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,13]],"date-time":"2026-05-13T13:58:51Z","timestamp":1778680731000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3779425"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,13]]},"references-count":39,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2026,5,31]]}},"alternative-id":["10.1145\/3779425"],"URL":"https:\/\/doi.org\/10.1145\/3779425","relation":{},"ISSN":["1539-9087","1558-3465"],"issn-type":[{"value":"1539-9087","type":"print"},{"value":"1558-3465","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,13]]},"assertion":[{"value":"2024-11-30","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-11-22","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-05-13","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}