{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,10]],"date-time":"2026-09-10T21:13:53Z","timestamp":1789074833736,"version":"build-2803163510"},"reference-count":53,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,10,19]],"date-time":"2025-10-19T00:00:00Z","timestamp":1760832000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,10,19]],"date-time":"2025-10-19T00:00:00Z","timestamp":1760832000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["W2412089"],"award-info":[{"award-number":["W2412089"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,10,19]]},"DOI":"10.1109\/iccv51701.2025.01395","type":"proceedings-article","created":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T19:45:49Z","timestamp":1777491949000},"page":"15035-15044","source":"Crossref","is-referenced-by-count":1,"title":["QuantCache: Adaptive Importance-Guided Quantization with Hierarchical Latent and Layer Caching for Video Generation"],"prefix":"10.1109","author":[{"given":"Junyi","family":"Wu","sequence":"first","affiliation":[{"name":"Shanghai Jiao Tong University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhiteng","family":"Li","sequence":"additional","affiliation":[{"name":"Shanghai Jiao Tong University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zheng","family":"Hui","sequence":"additional","affiliation":[{"name":"MGTV, Shanhai Academy"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yulun","family":"Zhang","sequence":"additional","affiliation":[{"name":"Shanghai Jiao Tong University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Linghe","family":"Kong","sequence":"additional","affiliation":[{"name":"Shanghai Jiao Tong University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaokang","family":"Yang","sequence":"additional","affiliation":[{"name":"Shanghai Jiao Tong University"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Quarot: Outlierfree 4-bit inference in rotated 11 ms","volume-title":"NeurIPS","author":"Ashkboos","year":"2024"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52734.2025.02636"},{"key":"ref3","author":"Chen","year":"2024","journal-title":"\u0394-dit: A training-free acceleration method tailored for diffusion transformers"},{"key":"ref4","article-title":"Analog bits: Generating discrete data using diffusion models with self-conditioning","volume-title":"ICLR","author":"Chen","year":"2023"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v39i15.33782"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/WACV61041.2025.00452"},{"key":"ref7","doi-asserted-by":"crossref","DOI":"10.18653\/v1\/2024.findings-emnlp.579","article-title":"Change is the only constant: Dynamic 11 m slicing based on layer redundancy","volume-title":"ACL","author":"Dumitru","year":"2024"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.01691"},{"key":"ref9","article-title":"What do vision transformers learn? a visual exploration","volume-title":"ICLR","author":"Ghiasi","year":"2023"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-73242-3_3"},{"key":"ref11","article-title":"Denoising diffusion probabilistic models","volume-title":"NeurIPS","author":"Ho","year":"2020"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02060"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02060"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.52202\/079017-4367"},{"key":"ref15","author":"Kahatapitiya","year":"2024","journal-title":"Adaptive caching for faster video generation with diffusion transformers"},{"key":"ref16","article-title":"Elucidating the design space of diffusion-based generative models","volume-title":"NeurIPS","author":"Karras","year":"2022"},{"key":"ref17","author":"Kim","year":"2024","journal-title":"A versatile diffusion transformer with mixture of noise levels for audiovisual generation"},{"key":"ref18","article-title":"Auto-encoding variational bayes","volume-title":"ICLR","author":"Kingma","year":"2014"},{"key":"ref19","article-title":"On density estimation with diffusion models","volume-title":"NeurIPS","author":"Kingma","year":"2021"},{"key":"ref20","volume-title":"Flux.","year":"2024"},{"key":"ref21","article-title":"Svdquant: Absorbing outliers by low-rank components for 4-bit diffusion models","volume-title":"ICLR","author":"Li","year":"2025"},{"key":"ref22","article-title":"Faster diffusion: Rethinking the role of the encoder for diffusion model inference","volume-title":"NeurIPS","author":"Li","year":"2024"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.01608"},{"key":"ref24","article-title":"S2phere: Semi-supervised pre-training for web search over heterogeneous learning to rank data","volume-title":"ACM SIGKDD","author":"Li","year":"2023"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1145\/3711896.3737258"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/ICDE65448.2025.00333"},{"key":"ref27","author":"Li","year":"2024","journal-title":"Hunyuan-dit: A powerful multi-resolution diffusion transformer with fine-grained chinese understanding"},{"key":"ref28","article-title":"Arb-llm: Alternating refined binarizations for large language models","volume-title":"ICLR","author":"Li","year":"2025"},{"key":"ref29","article-title":"Qserve: W4a8kv4 quantization and system co-design for efficient llm serving","volume-title":"MLSys","author":"Lin","year":"2025"},{"key":"ref30","author":"Lin","year":"2022","journal-title":"Genie: Large scale pre-training for text generation with diffusion model"},{"key":"ref31","article-title":"Faster diffusion via temporal attention decomposition","volume-title":"TMLR","author":"Liu","year":"2025"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02090"},{"key":"ref33","author":"Lu","year":"2024","journal-title":"Terdit: Ternary diffusion models with transformers"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01492"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.findings-acl.1035"},{"key":"ref36","article-title":"A white paper on neural network quantization","volume":"abs\/2106.08295","author":"Nagel","year":"2021","journal-title":"CoRR"},{"key":"ref37","article-title":"Improved denoising diffusion probabilistic models","volume-title":"ICCV","author":"Nichol","year":"2021"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.00387"},{"key":"ref39","article-title":"Consisti2v: Enhancing visual consistency for image-to-video generation","volume-title":"TMLR","author":"Ren","year":"2024"},{"key":"ref40","volume":"abs\/2202.00512","author":"Salimans","year":"2022","journal-title":"Progressive distillation for fast sampling of diffusion models"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.00196"},{"key":"ref42","article-title":"$\\text{Pb}-11 \\mathrm{m}$: Partially binarized large language models","volume-title":"ICLR","author":"Shang","year":"2024"},{"key":"ref43","article-title":"Denoising diffusion implicit models","volume-title":"ICLR","author":"Song","year":"2021"},{"key":"ref44","article-title":"Attention is all you need","volume-title":"NeurIPS","author":"Vaswani","year":"2017"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.01843"},{"key":"ref46","article-title":"Ptq4dit: Post-training quantization for diffusion transformers","volume-title":"NeurIPS","author":"Wu","year":"2025"},{"key":"ref47","article-title":"SmoothQuant: Accurate and efficient post-training quantization for large language models","volume-title":"ICML","author":"Xiao","year":"2023"},{"key":"ref48","article-title":"Cogvideox: Text-to-video diffusion models with an expert transformer","volume-title":"ICLR","author":"Yang","year":"2025"},{"key":"ref49","article-title":"Crossattention makes inference cumbersome in text-to-image diffusion models","volume-title":"TMLR","author":"Zhang","year":"2025"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v36i3.20252"},{"key":"ref51","article-title":"Vidit-q: Efficient and accurate quantization of diffusion transformers for image and video generation","volume-title":"ICLR","author":"Zhao","year":"2025"},{"key":"ref52","author":"Zhao","year":"2024","journal-title":"Real-time video generation with pyramid attention broadcast"},{"key":"ref53","volume-title":"Open-sora: Democratizing efficient video production for all","author":"Zheng","year":"2024"}],"event":{"name":"2025 IEEE\/CVF International Conference on Computer Vision (ICCV)","location":"Honolulu, HI, USA","start":{"date-parts":[[2025,10,19]]},"end":{"date-parts":[[2025,10,25]]}},"container-title":["2025 IEEE\/CVF International Conference on Computer Vision (ICCV)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11443115\/11443287\/11445489.pdf?arnumber=11445489","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,30]],"date-time":"2026-04-30T06:19:26Z","timestamp":1777529966000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11445489\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,19]]},"references-count":53,"URL":"https:\/\/doi.org\/10.1109\/iccv51701.2025.01395","relation":{},"subject":[],"published":{"date-parts":[[2025,10,19]]}}}