{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,4]],"date-time":"2026-05-04T03:30:57Z","timestamp":1777865457360,"version":"3.51.4"},"reference-count":66,"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\/iccv51701.2025.01826","type":"proceedings-article","created":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T19:45:49Z","timestamp":1777491949000},"page":"19638-19648","source":"Crossref","is-referenced-by-count":1,"title":["Multimodal LLMs as Customized Reward Models for Text-to-Image Generation"],"prefix":"10.1109","author":[{"given":"Shijie","family":"Zhou","sequence":"first","affiliation":[{"name":"University at Buffalo"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ruiyi","family":"Zhang","sequence":"additional","affiliation":[{"name":"Adobe Research"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Huaisheng","family":"Zhu","sequence":"additional","affiliation":[{"name":"Pennsylvania State University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Branislav","family":"Kveton","sequence":"additional","affiliation":[{"name":"Adobe Research"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yufan","family":"Zhou","sequence":"additional","affiliation":[{"name":"Luma AI"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jiuxiang","family":"Gu","sequence":"additional","affiliation":[{"name":"Adobe Research"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jian","family":"Chen","sequence":"additional","affiliation":[{"name":"University at Buffalo"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Changyou","family":"Chen","sequence":"additional","affiliation":[{"name":"University at Buffalo"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"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"},{"issue":"2","key":"ref2","article-title":"Qwen-vl: A versatile vision-language model for understanding, localization, text reading, and beyond","volume":"1","author":"Bai","year":"2023","journal-title":"arXiv preprint"},{"key":"ref3","article-title":"Qwen2. 5-vl technical report","author":"Bai","year":"2025","journal-title":"arXiv preprint"},{"key":"ref4","article-title":"A note on the inception score","author":"Barratt","year":"2018","journal-title":"arXiv preprint"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.blackboxnlp-1.10"},{"key":"ref6","article-title":"Training diffusion models with reinforcement learning","author":"Black","year":"2023","journal-title":"arXiv preprint"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.2307\/2334029"},{"key":"ref8","article-title":"Minigpt-v2: large language model as a unified interface for vision-language multi-task learning","author":"Chen","year":"2023","journal-title":"arXiv preprint"},{"key":"ref9","article-title":"Lora-contextualizing adaptation of large multimodal models for long document understanding","author":"Chen","year":"2024","journal-title":"arXiv preprint"},{"key":"ref10","article-title":"Mj-bench: Is your multimodal reward model really a good judge for text-toimage generation?","author":"Chen","year":"2024","journal-title":"arXiv preprint"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1007\/s11432-024-4231-5"},{"key":"ref12","article-title":"Damien Crone","volume-title":"The socio-moral image database (smid)","year":"2018"},{"key":"ref13","article-title":"Instructblip: Towards general-purpose vision-language models with instruction tuning. arxiv 2023","author":"Dai","year":"2023","journal-title":"arXiv preprint"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-main.798"},{"key":"ref15","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":"ref16","first-page":"19822","article-title":"Cogview: Mastering text-to-image generation via transformers","volume":"34","author":"Ding","year":"2021","journal-title":"Advances in neural information processing systems"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.52202\/075280-3497"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.52202\/075280-2270"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1145\/3422622"},{"key":"ref20","article-title":"Llavaguard: Vlm-based safeguards for vision dataset curation and safety assessment","author":"Helff","year":"2024","journal-title":"arXiv preprint"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2021.emnlp-main.595"},{"key":"ref22","article-title":"Gans trained by a two time-scale update rule converge to a local nash equilibrium","volume":"30","author":"Heusel","year":"2017","journal-title":"Advances in neural information processing systems"},{"key":"ref23","first-page":"6840","article-title":"Denoising diffusion probabilistic models","volume":"33","author":"Ho","year":"2020","journal-title":"Advances in neural information processing systems"},{"key":"ref24","article-title":"Lora: Low-rank adaptation of large language models","author":"Edward","year":"2021","journal-title":"arXiv preprint"},{"key":"ref25","article-title":"Llama guard: Llm-based input-output safeguard for human-ai conversations","author":"Inan","year":"2023","journal-title":"arXiv preprint"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.52202\/075280-1594"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.acl-long.663"},{"key":"ref28","article-title":"What matters when building vision-language models?","author":"Lauren\u00e7on","year":"2024","journal-title":"arXiv preprint"},{"key":"ref29","first-page":"1973019742","article-title":"Blip-2: Bootstrapping language-image pre-training with frozen image encoders and large language models","volume-title":"International conference on machine learning","author":"Li","year":"2023"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.01249"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72664-4_23"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72673-6_20"},{"key":"ref33","article-title":"Skywork-reward: Bag of tricks for reward modeling in 11 ms","author":"Yuhao Liu","year":"2024","journal-title":"arXiv preprint"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52733.2024.02484"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.52202\/075280-1516"},{"key":"ref36","author":"Liu","year":"2024","journal-title":"Llava-next: Improved reasoning, ocr, and world knowledge"},{"key":"ref37","author":"Liu","year":"2024","journal-title":"Llava-next: Improved reasoning, ocr, and world knowledge"},{"key":"ref38","article-title":"Inference-time scaling for diffusion models beyond scaling denoising steps","author":"Ma","year":"2025","journal-title":"arXiv preprint"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.52202\/079017-3946"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.52202\/068431-2011"},{"key":"ref41","article-title":"Sdxl: Improving latent diffusion models for high-resolution image synthesis","author":"Podell","year":"2023","journal-title":"arXiv preprint"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1145\/3576915.3616679"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1145\/3719027.3765088"},{"key":"ref44","first-page":"8748","article-title":"Learning transferable visual models from natural language supervision","volume-title":"International conference on machine learning","author":"Radford","year":"2021"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.52202\/075280-2338"},{"key":"ref46","article-title":"Direct preference optimization: Your language model is secretly a reward model","volume":"36","author":"Rafailov","year":"2024","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref47","first-page":"8821","article-title":"Zero-shot text-to-image generation","volume-title":"International conference on machine learning","author":"Ramesh","year":"2021"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01042"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.52202\/068431-2643"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.52202\/068431-1833"},{"key":"ref51","article-title":"Proximal policy optimization algorithms","author":"Schulman","year":"2017","journal-title":"arXiv preprint"},{"key":"ref52","article-title":"A general framework for inference-time scaling and steering of diffusion models","author":"Singhal","year":"2025","journal-title":"arXiv preprint"},{"key":"ref53","first-page":"2256","article-title":"Deep unsupervised learning using nonequilibrium thermodynamics","volume-title":"International conference on machine learning","author":"Sohl-Dickstein","year":"2015"},{"key":"ref54","article-title":"Rethinking bradley-terry models in preference-based reward modeling: Foundations, theory, and alternatives","author":"Sun","year":"2024","journal-title":"arXiv preprint"},{"key":"ref55","article-title":"Evalalign: Supervised finetuning multimodal 11 ms with human-aligned data for evaluating text-to-image models","author":"Tan","year":"2024","journal-title":"arXiv preprint"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.acl-long.1088"},{"key":"ref57","article-title":"Helpsteer2-preference: Complementing ratings with preferences","author":"Wang","year":"2024","journal-title":"arXiv preprint"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52734.2025.01366"},{"key":"ref59","article-title":"Q-align: Teaching 1 mms for visual scoring via discrete text-defined levels","author":"Wu","year":"2023","journal-title":"arXiv preprint"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.00200"},{"key":"ref61","article-title":"Imagereward: Learning and evaluating human preferences for text-to-image generation","volume":"36","author":"Xu","year":"2024","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2007.1078"},{"key":"ref63","article-title":"When and why visionlanguage models behave like bags-of-words, and what to do about it?","volume-title":"The Eleventh International Conference on Learning Representations","author":"Yuksekgonul","year":"2023"},{"key":"ref64","article-title":"General preference modeling with preference representations for aligning language models","author":"Zhang","year":"2024","journal-title":"arXiv preprint"},{"key":"ref65","article-title":"Treat visual tokens as text? but your mllm only needs fewer efforts to see","author":"Zhang","year":"2024","journal-title":"arXiv preprint"},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01738"}],"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\/11444114.pdf?arnumber=11444114","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,30]],"date-time":"2026-04-30T06:36:15Z","timestamp":1777530975000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11444114\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,19]]},"references-count":66,"URL":"https:\/\/doi.org\/10.1109\/iccv51701.2025.01826","relation":{},"subject":[],"published":{"date-parts":[[2025,10,19]]}}}