{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,8]],"date-time":"2026-10-08T19:07:14Z","timestamp":1791486434192,"version":"4.3.4"},"reference-count":81,"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\/iccv51701.2025.01713","type":"proceedings-article","created":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T19:45:49Z","timestamp":1777491949000},"page":"18431-18441","source":"Crossref","is-referenced-by-count":17,"title":["Randomized Autoregressive Visual Generation"],"prefix":"10.1109","author":[{"given":"Qihang","family":"Yu","sequence":"first","affiliation":[{"name":"ByteDance"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ju","family":"He","sequence":"additional","affiliation":[{"name":"ByteDance"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xueqing","family":"Deng","sequence":"additional","affiliation":[{"name":"ByteDance"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaohui","family":"Shen","sequence":"additional","affiliation":[{"name":"ByteDance"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Liang-Chieh","family":"Chen","sequence":"additional","affiliation":[{"name":"ByteDance"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Phi-3 technical report: A highly capable language model locally on your phone","author":"Abdin","year":"2024","journal-title":"arXiv preprint"},{"key":"ref2","article-title":"Gpt-4 technical report","author":"Achiam","year":"2023","journal-title":"arXiv preprint"},{"key":"ref3","article-title":"Palm 2 technical report","author":"Anil","year":"2023","journal-title":"arXiv preprint"},{"key":"ref4","article-title":"Qwen technical report","author":"Bai","year":"2023","journal-title":"arXiv preprint"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02157"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.02171"},{"key":"ref7","article-title":"Paligemma: A versatile 3b vlm for transfer","author":"Beyer","year":"2024","journal-title":"arXiv preprint"},{"key":"ref8","article-title":"Large scale gan training for high fidelity natural image synthesis","author":"Brock","year":"2018.3","journal-title":"arXiv preprint"},{"key":"ref9","article-title":"Language models are few-shot learners","volume-title":"NeurIPS","author":"Brown","year":"2020"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01103"},{"key":"ref11","article-title":"Muse: Text-to-image generation via masked generative transformers","volume-title":"ICML","author":"Chang","year":"2023"},{"key":"ref12","article-title":"Generative pretraining from pixels","volume-title":"ICML","author":"Chen","year":"2020"},{"issue":"240","key":"ref13","first-page":"1","article-title":"Palm: Scaling language modeling with pathways","volume":"24","author":"Chowdhery","year":"2023","journal-title":"Journal of Machine Learning Research"},{"issue":"70","key":"ref14","first-page":"1","article-title":"Scaling instructionfinetuned language models","volume":"25","author":"Chung","year":"2024","journal-title":"JMLR"},{"key":"ref15","first-page":"7480","article-title":"Scaling vision transformers to 22 billion parameters","volume-title":"ICML","author":"Dehghani","year":"2023"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref17","article-title":"Bert: Pre-training of deep bidirectional transformers for language understanding","volume-title":"NAACL","author":"Devlin","year":"2018"},{"key":"ref18","article-title":"Diffusion models beat gans on image synthesis","volume-title":"NeurIPS","author":"Dhariwal","year":"2021"},{"key":"ref19","article-title":"An image is worth 16x16 words: Transformers for image recognition at scale","volume-title":"ICLR","author":"Dosovitskiy","year":"2021"},{"key":"ref20","article-title":"The llama 3 herd of models","author":"Dubey","year":"2024","journal-title":"arXiv preprint"},{"key":"ref21","article-title":"Scalable pre-training of large autoregressive image models","volume-title":"ICML","author":"El-Nouby","year":"2024"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01268"},{"key":"ref23","article-title":"Scaling rectified flow transformers for high-resolution image synthesis","volume-title":"ICML","author":"Esser","year":"2024"},{"key":"ref24","article-title":"Fluid: Scaling autoregressive text-to-image generative models with continuous tokens","author":"Fan","year":"2024","journal-title":"arXiv preprint"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.02117"},{"key":"ref26","article-title":"Generative adversarial nets","volume-title":"NeurIPS","author":"Goodfellow","year":"2014"},{"key":"ref27","article-title":"Deep autoregressive networks","volume-title":"ICML","author":"Gregor","year":"2014"},{"key":"ref28","article-title":"Levenshtein transformer","volume-title":"NeurIPS","volume":"32","author":"Gu","year":"2019"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51701.2025.01544"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01553"},{"key":"ref31","article-title":"Classifier-free diffusion guidance","author":"Ho","year":"2022","journal-title":"arXiv preprint"},{"key":"ref32","article-title":"Denoising diffusion probabilistic models","volume-title":"NeurIPS","author":"Ho","year":"2020"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00453"},{"key":"ref34","article-title":"Democratizing text-to-image masked generative models with compact textaware one-dimensional tokens","volume-title":"ICCV","author":"Kim","year":"2025"},{"key":"ref35","article-title":"Adam: A method for stochastic optimization","volume-title":"ICLR","author":"Kingma","year":"2015"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1145\/3600006.3613165"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01123"},{"key":"ref38","article-title":"Autoregressive image generation without vector quantization","volume-title":"NeurIPS","author":"Li","year":"2024"},{"key":"ref39","article-title":"Alleviating distortion in image generation via multi-resolution diffusion models","volume-title":"NeurIPS","author":"Liu","year":"2024"},{"key":"ref40","article-title":"Decoupled weight decay regularization","volume-title":"ICLR","author":"Loshchilov","year":"2019"},{"key":"ref41","article-title":"Open-magvit2: An open-source project toward democratizing auto-regressive visual generation","author":"Luo","year":"2024","journal-title":"arXiv preprint"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72980-5_2"},{"key":"ref43","article-title":"Star: Scale-wise text-toimage generation via auto-regressive representations","author":"Ma","year":"2024","journal-title":"arXiv preprint"},{"key":"ref44","article-title":"Self-refine: Iterative refinement with self-feedback","volume-title":"NeurIPS","volume":"36","author":"Madaan","year":"2023"},{"key":"ref45","article-title":"Gpt-4 technical report","year":"2023","journal-title":"arXiv preprint"},{"key":"ref46","article-title":"Image transformer","volume-title":"ICML","author":"Parmar","year":"2018"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.00387"},{"key":"ref48","article-title":"Improving language understanding by generative pre-training","author":"Radford","year":"2018","journal-title":"OpenAI"},{"key":"ref49","article-title":"Learning transferable visual models from natural language supervision","volume-title":"ICML","author":"Radford","year":"2021"},{"key":"ref50","article-title":"Zero-shot text-to-image generation","volume-title":"ICML","author":"Ramesh","year":"2021"},{"key":"ref51","article-title":"Generating diverse high-fidelity images with vq-vae-2","volume-title":"NeurIPS","author":"Razavi","year":"2019"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51701.2025.01464"},{"key":"ref53","article-title":"Flowar: Scale-wise autoregressive image generation meets flow matching","volume-title":"ICML","author":"Ren","year":"2025"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01042"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51701.2025.01535"},{"key":"ref56","volume-title":"stabilityai","year":"2023"},{"key":"ref57","article-title":"Autoregressive model beats diffusion: Llama for scalable image generation","author":"Sun","year":"2024","journal-title":"arXiv preprint"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2015.7298594"},{"key":"ref59","article-title":"Hart: Efficient visual generation with hybrid autoregressive transformer","author":"Tang","year":"2024","journal-title":"arXiv preprint"},{"key":"ref60","article-title":"Chameleon: Mixed-modal early-fusion foundation models","year":"2024","journal-title":"arXiv preprint"},{"key":"ref61","article-title":"Emu3: Next-token prediction is all you need","year":"2024","journal-title":"Tech Report"},{"key":"ref62","article-title":"Gemini: a family of highly capable multimodal models","author":"Anil","year":"2023","journal-title":"arXiv preprint"},{"key":"ref63","article-title":"Visual autoregressive modeling: Scalable image generation via next-scale prediction","volume-title":"NeurIPS","author":"Tian","year":"2024"},{"key":"ref64","article-title":"Llama: Open and efficient foundation language models","author":"Touvron","year":"2023","journal-title":"arXiv preprint"},{"key":"ref65","article-title":"Llama 2: Open foundation and fine-tuned chat models","author":"Touvron","year":"2023","journal-title":"arXiv preprint"},{"issue":"205","key":"ref66","first-page":"1","article-title":"Neural autoregressive distribution estimation","volume":"17","author":"Uria","year":"2016","journal-title":"JMLR"},{"key":"ref67","article-title":"Conditional image generation with pixelcnn decoders","volume-title":"NeurIPS","author":"Van den Oord","year":"2016"},{"key":"ref68","article-title":"Pixel recurrent neural networks","volume-title":"ICML","author":"Van Den Oord","year":"2016"},{"key":"ref69","article-title":"Neural discrete representation learning","volume-title":"NeurIPS","author":"Van Den Oord","year":"2017"},{"key":"ref70","article-title":"Maskbit: Embedding-free image generation via bit tokens","author":"Weber","year":"2024","journal-title":"arXiv preprint"},{"key":"ref71","article-title":"Finetuned language models are zero-shot learners","volume-title":"ICLR","author":"Wei","year":"2022"},{"key":"ref72","article-title":"Qwen2 technical report","author":"Yang","year":"2024","journal-title":"arXiv preprint"},{"key":"ref73","article-title":"1.58-bit flux","author":"Yang","year":"2024","journal-title":"arXiv preprint"},{"key":"ref74","article-title":"Xlnet: Generalized autoregressive pretraining for language understanding","volume-title":"NeurIPS","author":"Yang","year":"2019"},{"key":"ref75","article-title":"Vector-quantized image modeling with improved vqgan","volume-title":"ICLR","author":"Yu","year":"2022"},{"key":"ref76","article-title":"Scaling autoregressive models for content-rich text-to-image generation","author":"Yu","year":"2022","journal-title":"TMLR"},{"key":"ref77","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01008"},{"key":"ref78","article-title":"Language model beats diffusion-tokenizer is key to visual generation","volume-title":"ICLR","author":"Yu","year":"2024"},{"key":"ref79","article-title":"An image is worth 32 tokens for reconstruction and generation","volume-title":"NeurIPS","author":"Yu","year":"2024"},{"key":"ref80","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01179"},{"key":"ref81","article-title":"Var-clip: Text-to-image generator with visual auto-regressive modeling","author":"Zhang","year":"2024","journal-title":"arXiv preprint"}],"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\/11444528.pdf?arnumber=11444528","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T04:53:32Z","timestamp":1777611212000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11444528\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,19]]},"references-count":81,"URL":"https:\/\/doi.org\/10.1109\/iccv51701.2025.01713","relation":{},"subject":[],"published":{"date-parts":[[2025,10,19]]}}}