{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,6]],"date-time":"2026-04-06T20:53:18Z","timestamp":1775508798105,"version":"3.50.1"},"reference-count":60,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"5","license":[{"start":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T00:00:00Z","timestamp":1777593600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T00:00:00Z","timestamp":1777593600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T00:00:00Z","timestamp":1777593600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"name":"Scientific Research Innovation Capability Support Project for Young Faculty","award":["ZYGXQNJSKYCXNLZCXM-I28"],"award-info":[{"award-number":["ZYGXQNJSKYCXNLZCXM-I28"]}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62476293"],"award-info":[{"award-number":["62476293"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62372482"],"award-info":[{"award-number":["62372482"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"name":"General Embodied AI Center of Sun Yat-sen University"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Pattern Anal. Mach. Intell."],"published-print":{"date-parts":[[2026,5]]},"DOI":"10.1109\/tpami.2026.3653573","type":"journal-article","created":{"date-parts":[[2026,1,13]],"date-time":"2026-01-13T20:58:38Z","timestamp":1768337918000},"page":"5725-5741","source":"Crossref","is-referenced-by-count":1,"title":["AtomThink: Multimodal Slow Thinking With Atomic Step Reasoning"],"prefix":"10.1109","volume":"48","author":[{"ORCID":"https:\/\/orcid.org\/0009-0006-5277-515X","authenticated-orcid":false,"given":"Kun","family":"Xiang","sequence":"first","affiliation":[{"name":"Shenzhen Campus of Sun Yat-sen University, Shenzhen, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhili","family":"Liu","sequence":"additional","affiliation":[{"name":"Hong Kong University of Science and Technology, Hong Kong, SAR, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-1271-4696","authenticated-orcid":false,"given":"Terry Jingchen","family":"Zhang","sequence":"additional","affiliation":[{"name":"ETH Zurich, Zurich, Switzerland"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0686-0832","authenticated-orcid":false,"given":"Yinya","family":"Huang","sequence":"additional","affiliation":[{"name":"ETH Zurich, Zurich, Switzerland"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-9717-2308","authenticated-orcid":false,"given":"Yunshuang","family":"Nie","sequence":"additional","affiliation":[{"name":"Shenzhen Campus of Sun Yat-sen University, Shenzhen, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Kaixin","family":"Cai","sequence":"additional","affiliation":[{"name":"Shenzhen Campus of Sun Yat-sen University, Shenzhen, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yiyang","family":"Yin","sequence":"additional","affiliation":[{"name":"Shenzhen Campus of Sun Yat-sen University, Shenzhen, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Runhui","family":"Huang","sequence":"additional","affiliation":[{"name":"University of Hong Kong, Hong Kong, SAR, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9838-6532","authenticated-orcid":false,"given":"Hanhui","family":"Li","sequence":"additional","affiliation":[{"name":"Shenzhen Campus of Sun Yat-sen University, Shenzhen, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-2441-5492","authenticated-orcid":false,"given":"Yihan","family":"Zeng","sequence":"additional","affiliation":[{"name":"Noah’s Ark Lab, Shanghai, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yu-Jie","family":"Yuan","sequence":"additional","affiliation":[{"name":"Noah’s Ark Lab, Shanghai, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jianhua","family":"Han","sequence":"additional","affiliation":[{"name":"Noah’s Ark Lab, Shanghai, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2752-5942","authenticated-orcid":false,"given":"Lanqing","family":"Hong","sequence":"additional","affiliation":[{"name":"Noah’s Ark Lab, Shanghai, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hang","family":"Xu","sequence":"additional","affiliation":[{"name":"Noah’s Ark Lab, Shanghai, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3213-3062","authenticated-orcid":false,"given":"Xiaodan","family":"Liang","sequence":"additional","affiliation":[{"name":"Shenzhen Campus of Sun Yat-sen University, Shenzhen, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.52202\/068431-1800"},{"key":"ref2","article-title":"OpenAI o1 system card"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-025-09422-z"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.65109\/hahs1010"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.52202\/075280-0517"},{"key":"ref6","article-title":"Interpretable contrastive Monte Carlo tree search reasoning","author":"Gao","year":"2024"},{"key":"ref7","article-title":"O1 replication journey: A strategic progress report\u2013part 1","author":"Qin","year":"2024"},{"key":"ref8","article-title":"Openr: An open source framework for advanced reasoning with large language models","author":"Wang","year":"2024"},{"key":"ref9","first-page":"2087","article-title":"Llava-o1: Let vision language models reason step-by-step","volume-title":"Proc. IEEE\/CVF Int. Conf. Comput. Vis.","author":"Xu","year":"2025"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.findings-acl.1247"},{"key":"ref11","article-title":"Do not think that much for 2 3=? on the overthinking of o1-like LLMs","author":"Chen","year":"2024"},{"key":"ref12","article-title":"Thoughts are all over the place: On the underthinking of o1-like LLMs","author":"Wang","year":"2025"},{"key":"ref13","first-page":"8309","article-title":"Towards VQA models that can read","volume-title":"Proc. IEEE\/CVF Conf. Comput. Vis. Pattern Recognit.","author":"Singh","year":"2019"},{"key":"ref14","article-title":"Learn to explain: Multimodal reasoning via thought chains for science question answering","author":"Lu","year":"2022"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1093\/nsr\/nwae403"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1145\/3773985"},{"key":"ref17","article-title":"Self-consistency improves chain of thought reasoning in language models","author":"Wang","year":"2022"},{"key":"ref18","article-title":"Let\u2019s verify step by step","volume-title":"Proc. 12th Int. Conf. Learn. Representations","author":"Lightman","year":"2023"},{"key":"ref19","article-title":"Quiet-Star: Language models can teach themselves to think before speaking","author":"Zelikman","year":"2024"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.acl-long.510"},{"key":"ref21","article-title":"Improve mathematical reasoning in language models by automated process supervision","author":"Luo","year":"2024"},{"key":"ref22","article-title":"Mavis: Mathematical visual instruction tuning with an automatic data engine","author":"Zhang","year":"2024"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.acl-long.82"},{"key":"ref24","article-title":"Scaling llm test-time compute optimally can be more effective than scaling model parameters","author":"Snell","year":"2024"},{"key":"ref25","article-title":"g1: Using Llama-3.1 70b on Groq to create o1-like reasoning chains","author":"Klieger","year":"2024"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.215"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2021.acl-long.528"},{"key":"ref28","article-title":"Dynamic Prompt Learning Via Policy Gradient for Semi-Structured Mathematical Reasoning","author":"Lu","year":"2022"},{"key":"ref29","article-title":"Geomverse: A systematic evaluation of large models for geometric reasoning","author":"Kazemi","year":"2023"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.findings-emnlp.268"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2021.findings-acl.46"},{"key":"ref32","article-title":"Iconqa: A new benchmark for abstract diagram understanding and visual language reasoning","author":"Lu","year":"2021"},{"key":"ref33","article-title":"R1-vision: Let\u2019s first take a look at the image","author":"Yu","year":"2025"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.52202\/075280-1516"},{"key":"ref35","article-title":"Llama 3.2: Revolutionizing edge AI and vision with open, customizable models","volume-title":"Meta"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.52202\/075280-2020"},{"key":"ref37","article-title":"The Llama 3 herd of models","author":"Grattafiori","year":"2024"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.acl-demos.38"},{"key":"ref39","article-title":"Claude 3.5 sonnet"},{"key":"ref40","article-title":"Gpt-4o system card"},{"key":"ref41","article-title":"Gpt-4v(ision) system card"},{"key":"ref42","article-title":"Llava-next: Improved reasoning, OCR, and world knowledge","author":"Liu","year":"2024"},{"key":"ref43","article-title":"Qwen technical report","author":"Bai","year":"2023"},{"key":"ref44","article-title":"Mathvista: Evaluating math reasoning in visual contexts with GPT-4V, bard, and other large multimodal models","author":"Lu","year":"2023","journal-title":"CoRR"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-73242-3_10"},{"key":"ref46","first-page":"95095","article-title":"Measuring multimodal mathematical reasoning with math-vision dataset","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"2024","author":"Wang"},{"key":"ref47","article-title":"We-math: Does your large multimodal model achieve human-like mathematical reasoning?","author":"Qiao","year":"2024"},{"key":"ref48","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-319-46493-0_15","article-title":"A diagram is worth a dozen images","author":"Kembhavi","year":"2016"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52733.2024.00913"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.70777\/si.v2i1.13973"},{"key":"ref51","first-page":"2199","article-title":"Docvqa: A dataset for VQA on document images","volume-title":"Proc. IEEE Winter Conf. Appl. Comput. Vis.","author":"Mathew","year":"2020"},{"key":"ref52","doi-asserted-by":"crossref","DOI":"10.18653\/v1\/2022.findings-acl.177","article-title":"ChartQA: A benchmark for question answering about charts with visual and logical reasoning","author":"Masry","year":"2022"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.findings-naacl.51"},{"key":"ref54","article-title":"LMMS-eval: Accelerating the development of large multimodal models","author":"Li","year":"2024"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.findings-acl.547"},{"key":"ref56","article-title":"Qwen2. 5-Math technical report: Toward mathematical expert model via self-improvement","author":"Yang","year":"2024"},{"key":"ref57","first-page":"8748","article-title":"Learning transferable visual models from natural language supervision","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Radford","year":"2021"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.00513"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52733.2024.02283"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01311"}],"container-title":["IEEE Transactions on Pattern Analysis and Machine Intelligence"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/34\/11474534\/11347549.pdf?arnumber=11347549","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,6]],"date-time":"2026-04-06T19:56:27Z","timestamp":1775505387000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11347549\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5]]},"references-count":60,"journal-issue":{"issue":"5"},"URL":"https:\/\/doi.org\/10.1109\/tpami.2026.3653573","relation":{},"ISSN":["0162-8828","2160-9292","1939-3539"],"issn-type":[{"value":"0162-8828","type":"print"},{"value":"2160-9292","type":"electronic"},{"value":"1939-3539","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5]]}}}