{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,12,11]],"date-time":"2025-12-11T15:12:58Z","timestamp":1765465978201,"version":"3.48.0"},"publisher-location":"New York, NY, USA","reference-count":47,"publisher":"ACM","funder":[{"name":"Semiconductor Research Corporation","award":[""],"award-info":[{"award-number":[""]}]},{"name":"AMD Gift","award":[""],"award-info":[{"award-number":[""]}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2026,3,22]]},"DOI":"10.1145\/3760250.3762220","type":"proceedings-article","created":{"date-parts":[[2025,12,11]],"date-time":"2025-12-11T15:06:36Z","timestamp":1765465596000},"page":"163-182","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["MoDM: Efficient Serving for Image Generation via\n                    <u>M<\/u>\n                    ixture-\n                    <u>o<\/u>\n                    f-\n                    <u>D<\/u>\n                    iffusion\n                    <u>M<\/u>\n                    odels"],"prefix":"10.1145","author":[{"ORCID":"https:\/\/orcid.org\/0009-0002-5557-3095","authenticated-orcid":false,"given":"Yuchen","family":"Xia","sequence":"first","affiliation":[{"name":"University of Michigan, Ann Arbor, MI, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-3740-7353","authenticated-orcid":false,"given":"Divyam","family":"Sharma","sequence":"additional","affiliation":[{"name":"University of Michigan, Ann Arbor, MI, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0509-8687","authenticated-orcid":false,"given":"Yichao","family":"Yuan","sequence":"additional","affiliation":[{"name":"University of Michigan, Ann Arbor, MI, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3533-9405","authenticated-orcid":false,"given":"Souvik","family":"Kundu","sequence":"additional","affiliation":[{"name":"Intel Labs, Los Angeles, CA, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2457-4119","authenticated-orcid":false,"given":"Nishil","family":"Talati","sequence":"additional","affiliation":[{"name":"University of Michigan, Ann Arbor, MI, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"320","published-online":{"date-parts":[[2025,12,11]]},"reference":[{"key":"e_1_3_2_1_1_1","unstructured":"Adobe unleashes new era of creativity for all with the commercial release of generative ai. 2023."},{"key":"e_1_3_2_1_2_1","unstructured":"Dall-e-2. 2023."},{"key":"e_1_3_2_1_3_1","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.acl-long.51"},{"key":"e_1_3_2_1_4_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.00594"},{"key":"e_1_3_2_1_5_1","volume-title":"Patchedserve: A patch management framework for slo-optimized hybrid resolution diffusion serving. arXiv preprint arXiv:2501.09253","author":"Sun Desen","year":"2025","unstructured":"Desen Sun, Zepeng Zhao, and Yuke Wang. Patchedserve: A patch management framework for slo-optimized hybrid resolution diffusion serving. arXiv preprint arXiv:2501.09253, 2025."},{"key":"e_1_3_2_1_6_1","doi-asserted-by":"publisher","DOI":"10.5555\/3691825.3691890"},{"key":"e_1_3_2_1_7_1","first-page":"15762","volume-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","author":"Ma Xinyin","year":"2024","unstructured":"Xinyin Ma, Gongfan Fang, and Xinchao Wang. Deepcache: Accelerating diffusion models for free. In Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 15762-15772, June 2024."},{"key":"e_1_3_2_1_8_1","first-page":"133282","article-title":"Accelerating diffusion transformer via layer caching","volume":"37","author":"Ma Xinyin","year":"2025","unstructured":"Xinyin Ma, Gongfan Fang, Michael Bi Mi, and Xinchao Wang. Learning-to-cache: Accelerating diffusion transformer via layer caching. Advances in Neural Information Processing Systems, 37:133282-133304, 2025.","journal-title":"Advances in Neural Information Processing Systems"},{"key":"e_1_3_2_1_9_1","volume-title":"Fora: Fast-forward caching in diffusion transformer acceleration. arXiv preprint arXiv:2407.01425","author":"Selvaraju Pratheba","year":"2024","unstructured":"Pratheba Selvaraju, Tianyu Ding, Tianyi Chen, Ilya Zharkov, and Luming Liang. Fora: Fast-forward caching in diffusion transformer acceleration. arXiv preprint arXiv:2407.01425, 2024."},{"key":"e_1_3_2_1_10_1","volume-title":"Accelerating diffusion transformers with dual feature caching. arXiv preprint arXiv:2412.18911","author":"Zou Chang","year":"2024","unstructured":"Chang Zou, Evelyn Zhang, Runlin Guo, Haohang Xu, Conghui He, Xuming Hu, and Linfeng Zhang. Accelerating diffusion transformers with dual feature caching. arXiv preprint arXiv:2412.18911, 2024."},{"key":"e_1_3_2_1_11_1","first-page":"288","volume-title":"European Conference on Computer Vision","author":"Lu Chen-Yi","year":"2024","unstructured":"Chen-Yi Lu, Shubham Agarwal, Md Mehrab Tanjim, Kanak Mahadik, Anup Rao, Subrata Mitra, Shiv Kumar Saini, Saurabh Bagchi, and Somali Chaterji. Recon: Training-free acceleration for text-to-image synthesis with retrieval of concept prompt trajectories. In European Conference on Computer Vision, pages 288-306. Springer, 2024."},{"key":"e_1_3_2_1_12_1","doi-asserted-by":"crossref","first-page":"7514","DOI":"10.18653\/v1\/2021.emnlp-main.595","volume-title":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing","author":"Hessel Jack","year":"2021","unstructured":"Jack Hessel, Ari Holtzman, Maxwell Forbes, Ronan Le Bras, and Yejin Choi. CLIPScore: A reference-free evaluation metric for image captioning. In Marie-Francine Moens, Xuanjing Huang, Lucia Specia, and Scott Wen-tau Yih, editors, Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 7514-7528, Online and Punta Cana, Dominican Republic, November 2021. Association for Computational Linguistics."},{"key":"e_1_3_2_1_13_1","volume-title":"Gans trained by a two time-scale update rule converge to a local nash equilibrium. Advances in neural information processing systems, 30","author":"Heusel Martin","year":"2017","unstructured":"Martin Heusel, Hubert Ramsauer, Thomas Unterthiner, Bernhard Nessler, and Sepp Hochreiter. Gans trained by a two time-scale update rule converge to a local nash equilibrium. Advances in neural information processing systems, 30, 2017."},{"key":"e_1_3_2_1_14_1","volume-title":"Scaling rectified flow transformers for high-resolution image synthesis","author":"Esser Patrick","year":"2024","unstructured":"Patrick Esser, Sumith Kulal, Andreas Blattmann, Rahim Entezari, Jonas M\u00fcller, Harry Saini, Yam Levi, Dominik Lorenz, Axel Sauer, Frederic Boesel, Dustin Podell, Tim Dockhorn, Zion English, Kyle Lacey, Alex Goodwin, Yannik Marek, and Robin Rombach. Scaling rectified flow transformers for high-resolution image synthesis, 2024."},{"key":"e_1_3_2_1_15_1","volume-title":"et al. Sana: Efficient high-resolution image synthesis with linear diffusion transformers. arXiv preprint arXiv:2410.10629","author":"Xie Enze","year":"2024","unstructured":"Enze Xie, Junsong Chen, Junyu Chen, Han Cai, Haotian Tang, Yujun Lin, Zhekai Zhang, Muyang Li, Ligeng Zhu, Yao Lu, et al. Sana: Efficient high-resolution image synthesis with linear diffusion transformers. arXiv preprint arXiv:2410.10629, 2024."},{"key":"e_1_3_2_1_16_1","volume-title":"High-resolution image synthesis with latent diffusion models","author":"Rombach Robin","year":"2021","unstructured":"Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Bj\u00f6rn Ommer. High-resolution image synthesis with latent diffusion models, 2021."},{"key":"e_1_3_2_1_17_1","unstructured":"Chitwan Saharia William Chan Saurabh Saxena Lala Li Jay Whang Emily L Denton Kamyar Ghasemipour Raphael Gontijo Lopes Burcu Karagol Ayan Tim Salimans et al. Photorealistic text-to-image diffusion models with deep language understanding. Advances in neural information processing systems 35:36479-36494 2022."},{"key":"e_1_3_2_1_18_1","volume-title":"Generative adversarial nets. Advances in neural information processing systems, 27","author":"Goodfellow Ian","year":"2014","unstructured":"Ian Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. Generative adversarial nets. Advances in neural information processing systems, 27, 2014."},{"key":"e_1_3_2_1_19_1","volume-title":"Auto-encoding variational bayes","author":"Kingma Diederik P","year":"2013","unstructured":"Diederik P Kingma, Max Welling, et al. Auto-encoding variational bayes, 2013."},{"key":"e_1_3_2_1_20_1","first-page":"36652","article-title":"An open dataset of user preferences for text-to-image generation","volume":"36","author":"Kirstain Yuval","year":"2023","unstructured":"Yuval Kirstain, Adam Polyak, Uriel Singer, Shahbuland Matiana, Joe Penna, and Omer Levy. Pick-a-pic: An open dataset of user preferences for text-to-image generation. Advances in Neural Information Processing Systems, 36:36652-36663, 2023.","journal-title":"Advances in Neural Information Processing Systems"},{"key":"e_1_3_2_1_21_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.308"},{"key":"e_1_3_2_1_22_1","volume-title":"Imagen 3. arXiv preprint arXiv:2408.07009","author":"Baldridge Jason","year":"2024","unstructured":"Jason Baldridge, Jakob Bauer, Mukul Bhutani, Nicole Brichtova, Andrew Bunner, Lluis Castrejon, Kelvin Chan, Yichang Chen, Sander Dieleman, Yuqing Du, et al. Imagen 3. arXiv preprint arXiv:2408.07009, 2024."},{"key":"e_1_3_2_1_23_1","volume-title":"The Twelfth International Conference on Learning Representations.","author":"Podell Dustin","unstructured":"Dustin Podell, Zion English, Kyle Lacey, Andreas Blattmann, Tim Dockhorn, Jonas M\u00fcller, Joe Penna, and Robin Rombach. Sdxl: Improving latent diffusion models for high-resolution image synthesis. In The Twelfth International Conference on Learning Representations."},{"key":"e_1_3_2_1_24_1","volume-title":"Adaptive training meets progressive scaling: Elevating efficiency in diffusion models. arXiv preprint arXiv:2312.13307","author":"Li Wenhao","year":"2023","unstructured":"Wenhao Li, Xiu Su, Yu Han, Shan You, Tao Huang, and Chang Xu. Adaptive training meets progressive scaling: Elevating efficiency in diffusion models. arXiv preprint arXiv:2312.13307, 2023."},{"key":"e_1_3_2_1_25_1","volume-title":"Denoising diffusion step-aware models. arXiv preprint arXiv:2310.03337","author":"Yang Shuai","year":"2023","unstructured":"Shuai Yang, Yukang Chen, Luozhou Wang, Shu Liu, and Yingcong Chen. Denoising diffusion step-aware models. arXiv preprint arXiv:2310.03337, 2023."},{"key":"e_1_3_2_1_26_1","volume-title":"Understanding the role of noise schedules in diffusion models. arXiv preprint arXiv:2206.00364","author":"Nichol Alex","year":"2022","unstructured":"Alex Nichol and Prafulla Dhariwal. Understanding the role of noise schedules in diffusion models. arXiv preprint arXiv:2206.00364, 2022."},{"key":"e_1_3_2_1_27_1","volume-title":"Early structure, late refinement: Understanding the progression of denoising in diffusion models. arXiv preprint arXiv:2305.01234","author":"Y. Li","year":"2023","unstructured":"Y. Li et al. Early structure, late refinement: Understanding the progression of denoising in diffusion models. arXiv preprint arXiv:2305.01234, 2023."},{"key":"e_1_3_2_1_28_1","doi-asserted-by":"publisher","DOI":"10.1145\/3620665.3640366"},{"key":"e_1_3_2_1_29_1","first-page":"219","article-title":"Pytorch rpc: Distributed deep learning built on tensor-optimized remote procedure calls","volume":"5","author":"Damania Pritam","year":"2023","unstructured":"Pritam Damania, Shen Li, Alban Desmaison, Alisson Azzolini, Brian Vaughan, Edward Yang, Gregory Chanan, Guoqiang Jerry Chen, Hongyi Jia, Howard Huang, et al. Pytorch rpc: Distributed deep learning built on tensor-optimized remote procedure calls. Proceedings of Machine Learning and Systems, 5:219-231, 2023.","journal-title":"Proceedings of Machine Learning and Systems"},{"key":"e_1_3_2_1_30_1","volume-title":"November","author":"Labs Black Forest","year":"2024","unstructured":"Black Forest Labs. FLUXkern0.1em.1: Introducing flux1 tools. https:\/\/blackforestlabs.io\/flux-1\/, November 2024. Accessed: 2025-07-22."},{"key":"e_1_3_2_1_31_1","unstructured":"PlaygroundAI. MJHQ-30K Dataset. 2023."},{"key":"e_1_3_2_1_32_1","volume-title":"Making stable diffusion faster with intelligent caching. https:\/\/www.pinecone.io\/learn\/faster-stable-diffusion\/","year":"2023","unstructured":"Pinecone. Making stable diffusion faster with intelligent caching. https:\/\/www.pinecone.io\/learn\/faster-stable-diffusion\/, 2023. Accessed: 2025-07-21."},{"key":"e_1_3_2_1_33_1","volume-title":"Sdedit: Guided image synthesis and editing with stochastic differential equations. arXiv preprint arXiv:2108.01073","author":"Meng Chenlin","year":"2021","unstructured":"Chenlin Meng, Yutong He, Yang Song, Jiaming Song, Jiajun Wu, Jun-Yan Zhu, and Stefano Ermon. Sdedit: Guided image synthesis and editing with stochastic differential equations. arXiv preprint arXiv:2108.01073, 2021."},{"key":"e_1_3_2_1_34_1","volume-title":"Echolm: Accelerating llm serving with real-time knowledge distillation. arXiv preprint arXiv:2501.12689","author":"Yu Yifan","year":"2025","unstructured":"Yifan Yu, Yu Gan, Lillian Tsai, Nikhil Sarda, Jiaming Shen, Yanqi Zhou, Arvind Krishnamurthy, Fan Lai, Henry M Levy, and David Culler. Echolm: Accelerating llm serving with real-time knowledge distillation. arXiv preprint arXiv:2501.12689, 2025."},{"key":"e_1_3_2_1_35_1","volume-title":"Progressive distillation for fast sampling of diffusion models. arXiv preprint arXiv:2202.00512","author":"Salimans Tim","year":"2022","unstructured":"Tim Salimans and Jonathan Ho. Progressive distillation for fast sampling of diffusion models. arXiv preprint arXiv:2202.00512, 2022."},{"key":"e_1_3_2_1_36_1","unstructured":"Gongfan Fang Xinyin Ma Xinchao Wang. Structural pruning for diffusion models\u2014supplementary materials\u2014."},{"key":"e_1_3_2_1_37_1","first-page":"17535","volume-title":"Proceedings of the IEEE\/CVF International Conference on Computer Vision","author":"Li Xiuyu","year":"2023","unstructured":"Xiuyu Li, Yijiang Liu, Long Lian, Huanrui Yang, Zhen Dong, Daniel Kang, Shanghang Zhang, and Kurt Keutzer. Q-diffusion: Quantizing diffusion models. In Proceedings of the IEEE\/CVF International Conference on Computer Vision, pages 17535-17545, 2023."},{"key":"e_1_3_2_1_38_1","volume-title":"Q-dm: An efficient low-bit quantized diffusion model. Advances in neural information processing systems, 36:76680-76691","author":"Li Yanjing","year":"2023","unstructured":"Yanjing Li, Sheng Xu, Xianbin Cao, Xiao Sun, and Baochang Zhang. Q-dm: An efficient low-bit quantized diffusion model. Advances in neural information processing systems, 36:76680-76691, 2023."},{"key":"e_1_3_2_1_39_1","volume-title":"Patch diffusion: Faster and more data-efficient training of diffusion models. Advances in neural information processing systems, 36:72137-72154","author":"Wang Zhendong","year":"2023","unstructured":"Zhendong Wang, Yifan Jiang, Huangjie Zheng, Peihao Wang, Pengcheng He, Zhangyang Wang, Weizhu Chen, Mingyuan Zhou, et al. Patch diffusion: Faster and more data-efficient training of diffusion models. Advances in neural information processing systems, 36:72137-72154, 2023."},{"issue":"120","key":"e_1_3_2_1_40_1","first-page":"1","article-title":"Scaling to trillion parameter models with simple and efficient sparsity","volume":"23","author":"Fedus William","year":"2022","unstructured":"William Fedus, Barret Zoph, and Noam Shazeer. Switch transformers: Scaling to trillion parameter models with simple and efficient sparsity. Journal of Machine Learning Research, 23(120):1-39, 2022.","journal-title":"Journal of Machine Learning Research"},{"key":"e_1_3_2_1_41_1","volume-title":"Heli Ben-Hamu, Maximilian Nickel, and Matt Le. Flow matching for generative modeling. arXiv preprint arXiv:2210.02747","author":"Lipman Yaron","year":"2022","unstructured":"Yaron Lipman, Ricky TQ Chen, Heli Ben-Hamu, Maximilian Nickel, and Matt Le. Flow matching for generative modeling. arXiv preprint arXiv:2210.02747, 2022."},{"key":"e_1_3_2_1_42_1","first-page":"9914","volume-title":"Adadiff: Adaptive step selection for fast diffusion models","author":"Zhang Hui","year":"2025","unstructured":"Hui Zhang, Zuxuan Wu, Zhen Xing, Jie Shao, and Yu-Gang Jiang. Adadiff: Adaptive step selection for fast diffusion models. In Toby Walsh, Julie Shah, and Zico Kolter, editors, AAAI-25, Sponsored by the Association for the Advancement of Artificial Intelligence, February 25 - March 4, 2025, Philadelphia, PA, USA, pages 9914-9922. AAAI Press, 2025."},{"key":"e_1_3_2_1_43_1","volume-title":"ICML 2023 Workshop on Structured Probabilistic Inference  Generative Modeling","author":"Moon Taehong","year":"2023","unstructured":"Taehong Moon, Moonseok Choi, EungGu Yun, Jongmin Yoon, Gayoung Lee, and Juho Lee. Early exiting for accelerated inference in diffusion models. In ICML 2023 Workshop on Structured Probabilistic Inference Generative Modeling, 2023."},{"key":"e_1_3_2_1_44_1","volume-title":"On fast sampling of diffusion probabilistic models. arXiv preprint arXiv:2106.00132","author":"Kong Zhifeng","year":"2021","unstructured":"Zhifeng Kong and Wei Ping. On fast sampling of diffusion probabilistic models. arXiv preprint arXiv:2106.00132, 2021."},{"key":"e_1_3_2_1_45_1","first-page":"8995","volume-title":"Proceedings of the IEEE\/CVF conference on computer vision and pattern Recognition","author":"Jiang Zeyinzi","year":"2024","unstructured":"Zeyinzi Jiang, Chaojie Mao, Yulin Pan, Zhen Han, and Jingfeng Zhang. Scedit: Efficient and controllable image diffusion generation via skip connection editing. In Proceedings of the IEEE\/CVF conference on computer vision and pattern Recognition, pages 8995-9004, 2024."},{"key":"e_1_3_2_1_46_1","volume-title":"Eighth Conference on Machine Learning and Systems","author":"Ahmad Sohaib","year":"2025","unstructured":"Sohaib Ahmad, Qizheng Yang, Haoliang Wang, Ramesh K. Sitaraman, and Hui Guan. Diffserve: Efficiently serving text-to-image diffusion models with query-aware model scaling. In Eighth Conference on Machine Learning and Systems, 2025."},{"key":"e_1_3_2_1_47_1","volume-title":"USENIX NSDI","author":"You Jie","year":"2023","unstructured":"Jie You, Jae-Won Chung, and Mosharaf Chowdhury. Zeus: Understanding and optimizing GPU energy consumption of DNN training. In USENIX NSDI, 2023."}],"event":{"name":"ASPLOS '26:31st ACM International Conference on Architectural Support for Programming Languages and Operating Systems","location":"Pittsburgh PA USA","sponsor":["SIGOPS ACM Special Interest Group on Operating Systems","SIGPLAN ACM Special Interest Group on Programming Languages","SIGARCH ACM Special Interest Group on Computer Architecture","SIGBED ACM Special Interest Group on Embedded Systems"]},"container-title":["Proceedings of the 31st ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 1"],"original-title":[],"link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3760250.3762220","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,11]],"date-time":"2025-12-11T15:08:06Z","timestamp":1765465686000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3760250.3762220"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,12,11]]},"references-count":47,"alternative-id":["10.1145\/3760250.3762220","10.1145\/3760250"],"URL":"https:\/\/doi.org\/10.1145\/3760250.3762220","relation":{},"subject":[],"published":{"date-parts":[[2025,12,11]]},"assertion":[{"value":"2025-12-11","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}