{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,4]],"date-time":"2026-05-04T09:56:38Z","timestamp":1777888598633,"version":"3.51.4"},"reference-count":62,"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"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,10,19]]},"DOI":"10.1109\/iccv51701.2025.01571","type":"proceedings-article","created":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T19:45:49Z","timestamp":1777491949000},"page":"16916-16926","source":"Crossref","is-referenced-by-count":0,"title":["Holistic Tokenizer for Autoregressive Image Generation"],"prefix":"10.1109","author":[{"given":"Anlin","family":"Zheng","sequence":"first","affiliation":[{"name":"The University of Hong Kong"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Haochen","family":"Wang","sequence":"additional","affiliation":[{"name":"NLPR, MAIS, CASIA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yucheng","family":"Zhao","sequence":"additional","affiliation":[{"name":"Dexmal"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Weipeng","family":"Deng","sequence":"additional","affiliation":[{"name":"The University of Hong Kong"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Tiancai","family":"Wang","sequence":"additional","affiliation":[{"name":"Dexmal"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xiangyu","family":"Zhang","sequence":"additional","affiliation":[{"name":"StepFun"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xiaojuan","family":"Qi","sequence":"additional","affiliation":[{"name":"The University of Hong Kong"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","author":"Bai","year":"2023","journal-title":"Qwen technical report"},{"key":"ref2","author":"Bengio","year":"2013","journal-title":"Estimating or propagating gradients through stochastic neurons for conditional computation"},{"key":"ref3","author":"Beyer","year":"2024","journal-title":"Paligemma: A versatile 3b vlm for transfer"},{"key":"ref4","author":"Brown","journal-title":"Language models are few-shot learners"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01103"},{"key":"ref6","first-page":"1691","article-title":"Generative pretraining from pixels","volume-title":"International conference on machine learning","author":"Chen","year":"2020"},{"key":"ref7","author":"Darcet","year":"2023","journal-title":"Vision transformers need registers"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref9","author":"Devlin","year":"2018","journal-title":"Bert: Pre-training of deep bidirectional transformers for language understanding"},{"key":"ref10","first-page":"8780","article-title":"Diffusion models beat gans on image synthesis","volume":"34","author":"Dhariwal","year":"2021","journal-title":"Advances in neural information processing systems"},{"key":"ref11","author":"El-Nouby","year":"2024","journal-title":"Scalable pretraining of large autoregressive image models"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01268"},{"key":"ref13","article-title":"Google","year":"2023","journal-title":"Bard"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01553"},{"key":"ref16","author":"Henighan","journal-title":"Scaling laws for autoregressive generative modeling"},{"key":"ref17","article-title":"Gans trained by a two time-scale update rule converge to a local nash equilibrium","author":"Heusel","year":"2017","journal-title":"Advances in neural information processing systems, 30"},{"key":"ref18","author":"Ho","year":"2022","journal-title":"Classifier-free diffusion guidance"},{"issue":"47","key":"ref19","first-page":"1","article-title":"Cascaded diffusion models for high fidelity image generation","volume":"23","author":"Ho","year":"2022","journal-title":"Journal of Machine Learning Research"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.632"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.00976"},{"key":"ref22","author":"Kaplan","journal-title":"Scaling laws for neural language models"},{"key":"ref23","author":"Kingma","year":"2013","journal-title":"Auto-encoding variational bayes"},{"key":"ref24","article-title":"Improved precision and recall metric for assessing generative models","author":"Kynk\u00e4\u00e4nniemi","year":"2019","journal-title":"Advances in neural information processing systems"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01123"},{"key":"ref26","author":"Li","year":"2024","journal-title":"Autoregressive image generation without vector quantization"},{"key":"ref27","first-page":"213","article-title":"Can ood object detectors learn from foundation models","volume-title":"European Conference on Computer Vision","author":"Liu","year":"2024"},{"key":"ref28","author":"Luo","year":"2024","journal-title":"Open-magvit2: An open-source project toward democratizing auto-regressive visual generation"},{"key":"ref29","author":"Mentzer","year":"2023","journal-title":"Finite scalar quantization: Vq-vae made simple"},{"key":"ref30","article-title":"OpenAI","year":"2022","journal-title":"Chatgpt"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.00387"},{"key":"ref32","author":"Podell","year":"2023","journal-title":"Sdxl: Improving latent diffusion models for high-resolution image synthesis"},{"key":"ref33","author":"Radford","year":"2018","journal-title":"Improving language understanding by generative pre-training"},{"issue":"8","key":"ref34","article-title":"Language models are unsupervised multitask learners","volume":"1","author":"Radford","year":"2019","journal-title":"OpenAI blog"},{"key":"ref35","first-page":"8748","article-title":"Learning transferable visual models from natural language supervision","volume-title":"International conference on machine learning","author":"Radford","year":"2021"},{"issue":"140","key":"ref36","first-page":"1","article-title":"Exploring the limits of transfer learning with a unified text-to-text transformer","volume":"21","author":"Raffel","year":"2020","journal-title":"Journal of machine learning research"},{"key":"ref37","first-page":"8821","article-title":"Zero-shot text-to-image generation","volume-title":"International conference on machine learning","author":"Ramesh","year":"2021"},{"key":"ref38","article-title":"Generating diverse high-fidelity images with vq-vae-2","author":"Razavi","year":"2019","journal-title":"Advances in neural information processing systems, 32"},{"key":"ref39","article-title":"Generating diverse high-fidelity images with vq-vae-2","author":"Razavi","year":"2019","journal-title":"Advances in neural information processing systems, 32"},{"key":"ref40","author":"Rombach","year":"2021","journal-title":"High-resolution image synthesis with latent diffusion models"},{"key":"ref41","article-title":"Improved techniques for training gans","author":"Salimans","year":"2016","journal-title":"Advances in neural information processing systems, 29"},{"key":"ref42","first-page":"1","article-title":"Scaling stylegan to large diverse datasets","volume-title":"Proceedings of the SIGGRAPH Conference","author":"Sauer"},{"key":"ref43","author":"Song","journal-title":"Denoising diffusion implicit models"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2023.127063"},{"key":"ref45","author":"Sun","year":"2024","journal-title":"Autoregressive model beats diffusion: Llama for scalable image generation"},{"key":"ref46","author":"Tian","year":"2024","journal-title":"Visual autoregressive modeling: Scalable image generation via next-scale prediction"},{"key":"ref47","author":"Touvron","year":"2023","journal-title":"Llama: Open and efficient foundation language models"},{"key":"ref48","article-title":"Conditional image generation with pixelcnn decoders","volume":"29","author":"Van den Oord","year":"2016","journal-title":"Advances in neural information processing systems"},{"key":"ref49","article-title":"Neural discrete representation learning","author":"Van","year":"2017","journal-title":"Advances in neural information processing systems"},{"key":"ref50","article-title":"Attention is all you need","author":"Vaswani","year":"2017","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref51","article-title":"Reconstructive visual instruction tuning","volume-title":"International Conference on Learning Representations","author":"Wang","year":"2025"},{"key":"ref52","author":"Xie","year":"2024","journal-title":"Show-o: One single transformer to unify multimodal understanding and generation"},{"key":"ref53","author":"Yu","year":"2021","journal-title":"Vector-quantized image modeling with improved vqgan"},{"key":"ref54","author":"Yu","year":"2021","journal-title":"Vector-quantized image modeling with improved vqgan"},{"issue":"3","key":"ref55","volume":"2","author":"Yu","year":"2022","journal-title":"Scaling autoregressive models for content-rich text-to-image generation"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01008"},{"key":"ref57","author":"Yu","year":"2023","journal-title":"Language model beats diffusion-tokenizer is key to visual generation"},{"key":"ref58","author":"Yu","year":"2024","journal-title":"An image is worth 32 tokens for reconstruction and generation"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00068"},{"key":"ref60","author":"Zhang","year":"2022","journal-title":"Opt: Open pre-trained transformer language models"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.52202\/068431-1701"},{"key":"ref62","author":"Zhu","year":"2024","journal-title":"Scaling the codebook size of vqgan to 100,000 with a utilization rate of 99%"}],"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\/11444323.pdf?arnumber=11444323","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T04:51:21Z","timestamp":1777611081000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11444323\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,19]]},"references-count":62,"URL":"https:\/\/doi.org\/10.1109\/iccv51701.2025.01571","relation":{},"subject":[],"published":{"date-parts":[[2025,10,19]]}}}