{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,4]],"date-time":"2026-05-04T10:21:34Z","timestamp":1777890094669,"version":"3.51.4"},"reference-count":73,"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\/iccv51701.2025.00327","type":"proceedings-article","created":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T19:45:49Z","timestamp":1777491949000},"page":"3424-3434","source":"Crossref","is-referenced-by-count":0,"title":["Enhancing Numerical Prediction of MLLMS With Soft Labeling"],"prefix":"10.1109","author":[{"given":"Pei","family":"Wang","sequence":"first","affiliation":[{"name":"Amazon AGI"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhaowei","family":"Cai","sequence":"additional","affiliation":[{"name":"Amazon AGI"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hao","family":"Yang","sequence":"additional","affiliation":[{"name":"Amazon AGI"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Davide","family":"Modolo","sequence":"additional","affiliation":[{"name":"Amazon AGI"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ashwin","family":"Swaminathan","sequence":"additional","affiliation":[{"name":"Amazon AGI"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v33i01.33018076"},{"key":"ref2","article-title":"Gpt-4 technical report","author":"Achiam","year":"2023","journal-title":"arXiv preprint"},{"key":"ref3","article-title":"Screenai: A vision-language model for ui and infographics understanding","author":"Baechler","year":"2024","journal-title":"arXiv preprint"},{"issue":"8","key":"ref4","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 2023. arXiv preprint arXiv"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2021.115271"},{"key":"ref6","first-page":"411","article-title":"Unimodal probability distributions for deep ordinal classification","volume-title":"In ICML","author":"Beckham","year":"2017"},{"issue":"18771901","key":"ref7","article-title":"Language models are few-shot learners","volume":"33","author":"Brown","year":"2020","journal-title":"NeurIPS"},{"key":"ref8","article-title":"Chart-based reasoning: Transferring capabilities from 11 ms to vlms","author":"Carbune","year":"2024","journal-title":"arXiv preprint"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/tnnls.2022.3179355"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02506"},{"key":"ref11","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":"ref12","article-title":"Shikra: Unleashing multimodal llm\u2019s referential dialogue magic","author":"Chen","year":"2023","journal-title":"arXiv preprint"},{"key":"ref13","first-page":"24185","article-title":"Internvl: Scaling up vision foundation models and aligning for generic visual-linguistic tasks","author":"Chen","year":"2024","journal-title":"In CVPR"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.acl-long.66"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00487"},{"key":"ref16","article-title":"Test of time: A benchmark for evaluating 11 ms on temporal reasoning","author":"Fatemi","year":"2024","journal-title":"arXiv preprint"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00214"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.670"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/TKDE.2015.2457911"},{"key":"ref20","article-title":"Squared earth mover\u2019s distance-based loss for training deep neural networks","author":"Hou","year":"2016","journal-title":"arXiv preprint"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/TMM.2018.2866770"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.00916"},{"key":"ref23","article-title":"Gpt-4o system card","author":"Hurst","year":"2024","journal-title":"arXiv preprint"},{"key":"ref24","article-title":"Figureqa: An annotated figure dataset for visual reasoning","author":"Kahou","year":"2017","journal-title":"arXiv preprint"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.3115\/v1\/D14-1086"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D18-2012"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-020-01316-z"},{"key":"ref28","article-title":"Llava-onevision: Easy visual task transfer","author":"Li","year":"2024","journal-title":"arXiv preprint"},{"key":"ref29","first-page":"13613","article-title":"Synthesize step-by-step: Tools templates and 11 ms as data generators for reasoning-based chart vqa","author":"Li","year":"2024","journal-title":"In CVPR"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.00018"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.52202\/075280-1516"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02484"},{"key":"ref33","author":"Liu","year":"2024","journal-title":"Llava-next: Improved reasoning, ocr, and world knowledge"},{"key":"ref34","first-page":"2210","article-title":"Label smoothing for text mining","author":"Liu","year":"2022","journal-title":"In ICCL"},{"key":"ref35","first-page":"0","article-title":"Ordinal regression with neuron stick-breaking for medical diagnosis","volume-title":"In ECCV Workshops","author":"Liu","year":"2018"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2020.01.025"},{"key":"ref37","article-title":"Revisiting multi-modal 11 m evaluation","author":"Lu","year":"2024","journal-title":"arXiv preprint"},{"key":"ref38","article-title":"Mathvista: Evaluating mathematical reasoning of foundation models in visual contexts","author":"Lu","year":"2023","journal-title":"arXiv preprint"},{"key":"ref39","first-page":"6448","article-title":"Does label smoothing mitigate label noise?","volume-title":"ICML","author":"Lukasik","year":"2020"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2022.findings-acl.177"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-main.906"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/WACV48630.2021.00225"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/WACV45572.2020.9093523"},{"key":"ref44","volume":"32","author":"M\u00fcller","year":"2019","journal-title":"When does label smoothing help? NeurIPS"},{"key":"ref45","article-title":"Movie: Revisiting modulated convolutions for visual counting and beyond","author":"Nguyen","year":"2020","journal-title":"arXiv preprint"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.532"},{"key":"ref47","article-title":"Kosmos-2: Grounding multimodal large language models to the world","author":"Peng","year":"2023","journal-title":"arXiv preprint"},{"key":"ref48","article-title":"Grounding multimodal large language models to the world","author":"Peng","year":"2024","journal-title":"ICLR"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-12053-4_12"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01335"},{"key":"ref51","article-title":"Momentor: Advancing video large language model with fine-grained temporal reasoning","author":"Qian","year":"2024","journal-title":"arXiv preprint"},{"issue":"8","key":"ref52","first-page":"9","article-title":"Language models are unsupervised multitask learners","volume":"1","author":"Radford","year":"2019","journal-title":"OpenAI blog"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00340"},{"issue":"2527825294","key":"ref54","article-title":"Laion-5b: An open large-scale dataset for training next generation image-text models","volume":"35","author":"Schuhmann","year":"2022","journal-title":"NeurIPS"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00852"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00851"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.308"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-54187-7_14"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v35i15.17635"},{"key":"ref60","article-title":"Gemini: a family of highly capable multimodal models","author":"Team","year":"2023","journal-title":"arXiv preprint"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-540-85563-7_14"},{"key":"ref62","author":"Tong","year":"2024","journal-title":"Cambrian-1: A fully open, vision-centric exploration of multimodal 11 ms"},{"key":"ref63","article-title":"Llama: Open and efficient foundation language models","author":"Touvron","year":"2023","journal-title":"arXiv preprint"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.463"},{"key":"ref65","doi-asserted-by":"publisher","DOI":"10.1016\/j.inffus.2023.01.003"},{"key":"ref66","first-page":"468","article-title":"Geols: geodesic label smoothing for image segmentation","author":"Vasudeva","year":"2024","journal-title":"In Medical Imaging with Deep Learning"},{"key":"ref67","first-page":"95095","article-title":"Measuring multimodal mathematical reasoning with math-vision dataset","volume":"37","author":"Wang","year":"2024","journal-title":"NeurIPS"},{"key":"ref68","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.596"},{"key":"ref69","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01347"},{"key":"ref70","article-title":"Ferret: Refer and ground anything anywhere at any granularity","author":"You","year":"2023","journal-title":"arXiv preprint"},{"key":"ref71","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2021.3089942"},{"key":"ref72","article-title":"Ferretv 2: An improved baseline for referring and grounding with large language models","author":"Zhang","year":"2024","journal-title":"arXiv preprint"},{"key":"ref73","article-title":"Rethinking soft labels for knowledge distillation: A bias-variance tradeoff perspective","author":"Zhou","year":"2021","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\/11444577.pdf?arnumber=11444577","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T05:29:52Z","timestamp":1777613392000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11444577\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,19]]},"references-count":73,"URL":"https:\/\/doi.org\/10.1109\/iccv51701.2025.00327","relation":{},"subject":[],"published":{"date-parts":[[2025,10,19]]}}}