{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T15:34:50Z","timestamp":1782833690545,"version":"3.54.5"},"reference-count":64,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,6,30]],"date-time":"2025-06-30T00:00:00Z","timestamp":1751241600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,6,30]],"date-time":"2025-06-30T00:00:00Z","timestamp":1751241600000},"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":[[2025,6,30]]},"DOI":"10.1109\/ijcnn64981.2025.11228391","type":"proceedings-article","created":{"date-parts":[[2025,11,14]],"date-time":"2025-11-14T18:46:15Z","timestamp":1763145975000},"page":"1-8","source":"Crossref","is-referenced-by-count":2,"title":["Retrieval-Augmented Multi-Modal Chain-of-Thoughts Reasoning for Large Language Models"],"prefix":"10.1109","author":[{"given":"Bingshuai","family":"Liu","sequence":"first","affiliation":[{"name":"Xiamen University,School of Informatics,Xiamen,China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chenyang","family":"Lyu","sequence":"additional","affiliation":[{"name":"Alibaba International Digital Commerce"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zijun","family":"Min","sequence":"additional","affiliation":[{"name":"Xiamen University,School of Informatics,Xiamen,China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhanyu","family":"Wang","sequence":"additional","affiliation":[{"name":"Tiktok"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jinsong","family":"Su","sequence":"additional","affiliation":[{"name":"Xiamen University,School of Informatics,Xiamen,China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Longyue","family":"Wang","sequence":"additional","affiliation":[{"name":"Alibaba International Digital Commerce"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Chain of thought prompting elicits reasoning in large language models","author":"Wei","year":"2022"},{"key":"ref2","article-title":"Large language models are zero-shot reasoners","author":"Kojima","year":"2022"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.findings-emnlp.734"},{"key":"ref4","article-title":"Chameleon: Plug-and-play compositional reasoning with large language models","author":"Lu","year":"2023"},{"key":"ref5","article-title":"Macaw-llm: Multi-modal language modeling with image, audio, video, and text integration","author":"Lyu","year":"2023"},{"key":"ref6","article-title":"A comprehensive evaluation of gpt-4v on knowledge-intensive visual question answering","author":"Li","year":"2023"},{"key":"ref7","article-title":"Self-consistency improves chain of thought reasoning in language models","author":"Wang","year":"2022"},{"key":"ref8","article-title":"Least-to-most prompting enables complex reasoning in large language models","volume":"abs\/2205.10625","author":"Zhou","year":"2022","journal-title":"ArXiv preprint"},{"key":"ref9","article-title":"Automatic chain of thought prompting in large language models","volume":"abs\/2210.03493","author":"Zhang","year":"2022","journal-title":"CoRR"},{"key":"ref10","first-page":"25005","article-title":"Visual prompting via image inpainting","volume":"35","author":"Bar","year":"2022","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/tpami.2025.3571946"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/tpami.2025.3571946"},{"key":"ref13","article-title":"Learn to explain: Multimodal reasoning via thought chains for science question answering","volume-title":"The 36th Conference on Neural Information Processing Systems (NeurIPS)","author":"Lu"},{"key":"ref14","article-title":"Mathvista: Evaluating mathematical reasoning of foundation models in visual contexts","author":"Lu","year":"2023"},{"key":"ref15","article-title":"What makes good examples for visual in-context learning?","author":"Zhang","year":"2023"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/tip.2025.3554410"},{"key":"ref17","article-title":"Realm: Retrieval-augmented language model pre-training","volume":"abs\/2002.08909","author":"Guu","year":"2020"},{"key":"ref18","article-title":"Retrieval-augmented generation for knowledge-intensive nlp tasks","volume":"abs\/2005.11401","author":"Lewis","year":"2020"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2022.emnlp-main.375"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.02238"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1145\/3539618.3591687"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.naacl-long.463"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00605"},{"key":"ref24","article-title":"Pre-computed memory or on-the-fly encoding? a hybrid approach to retrieval augmentation makes the most of your compute","volume":"abs\/2301.10448","author":"de Jong","year":"2023"},{"key":"ref25","first-page":"4171","article-title":"BERT: Pre-training of deep bidirectional transformers for language understanding","volume-title":"Proc. of NAACL-HLT","author":"Devlin"},{"key":"ref26","article-title":"Language models are unsupervised multitask learners","author":"Radford","year":"2019"},{"key":"ref27","article-title":"Language models are few-shot learners","volume-title":"Advances in Neural Information Processing Systems 33: Annual Conference on Neural Information Processing Systems 2020, NeurIPS 2020, December 6-12, 2020, virtual","author":"Brown"},{"key":"ref28","article-title":"PaLM: Scaling language modeling with pathways","author":"Chowdhery","year":"2022"},{"key":"ref29","article-title":"Chain of thought prompting elicits reasoning in large language models","volume":"abs\/2201.11903","author":"Wei","year":"2022","journal-title":"ArXiv preprint"},{"key":"ref30","first-page":"12697","article-title":"Calibrate before use: Improving few-shot performance of language models","volume-title":"Proc. of ICML","author":"Zhao"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.acl-long.153"},{"key":"ref32","article-title":"Complexity-based prompting for multi-step reasoning","volume":"abs\/2210.00720","author":"Fu","year":"2022","journal-title":"CoRR"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.00660"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/iccv51070.2023.00110"},{"key":"ref35","article-title":"In-context learning unlocked for diffusion models","author":"Wang","year":"2023"},{"key":"ref36","first-page":"200","article-title":"Multimodal few-shot learning with frozen language models","volume":"34","author":"Tsimpoukelli","year":"2021","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref37","first-page":"23716","article-title":"Flamingo: a visual language model for few-shot learning","volume":"35","author":"Alayrac","year":"2022","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref38","article-title":"Prodigy: Enabling in-context learning over graphs","author":"Huang","year":"2023"},{"key":"ref39","article-title":"Language models are multilingual chain-of-thought reasoners","volume":"abs\/2210.03057","author":"Shi","year":"2022","journal-title":"ArXiv preprint"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.findings-acl.651"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.findings-emnlp.641"},{"key":"ref42","first-page":"2320","article-title":"Stratified sampling meets machine learning","volume-title":"Proceedings of The 33rd International Conference on Machine Learning","volume":"48","author":"Liberty"},{"key":"ref43","article-title":"Chatgpt","year":"2022"},{"key":"ref44","article-title":"GPT-4 technical report","volume":"abs\/2303.08774","year":"2023","journal-title":"ArXiv"},{"key":"ref45","article-title":"Visual instruction tuning","author":"Liu","year":"2023"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.nlp4science-1.7"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D19-1410"},{"key":"ref48","article-title":"An image is worth 16x16 words: Transformers for image recognition at scale","author":"Dosovitskiy","year":"2021"},{"key":"ref49","article-title":"Learning transferable visual models from natural language supervision","author":"Radford","year":"2021"},{"key":"ref50","article-title":"Blip-2: Bootstrapping language-image pre-training with frozen image encoders and large language models","author":"Li","year":"2023"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.3233\/faia240581"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1145\/3664647.3681464"},{"key":"ref53","first-page":"2023","article-title":"A comprehensive study of gpt-4v\u2019s multimodal capabilities in medical imaging","author":"Li","year":"2023","journal-title":"medRxiv"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1145\/3571730"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1162\/coli.a.16"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.findings-emnlp.262"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.acl-long.1167"},{"key":"ref58","article-title":"Litesearch: Efficacious tree search for llm","author":"Wang","year":"2024"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.findings-acl.865"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.findings-emnlp.740"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.1109\/tpami.2025.3532688"},{"key":"ref62","article-title":"Can multimodal llms do visual temporal understanding and reasoning? the answer is no!","author":"Imam","year":"2025"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP48485.2024.10447275"},{"key":"ref64","first-page":"11479","article-title":"Cvqa: Culturally-diverse multilingual visual question answering benchmark","volume":"37","author":"Romero","year":"2025","journal-title":"Advances in Neural Information Processing Systems"}],"event":{"name":"2025 International Joint Conference on Neural Networks (IJCNN)","location":"Rome, Italy","start":{"date-parts":[[2025,6,30]]},"end":{"date-parts":[[2025,7,5]]}},"container-title":["2025 International Joint Conference on Neural Networks (IJCNN)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11227166\/11227148\/11228391.pdf?arnumber=11228391","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,15]],"date-time":"2025-11-15T07:27:15Z","timestamp":1763191635000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11228391\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,6,30]]},"references-count":64,"URL":"https:\/\/doi.org\/10.1109\/ijcnn64981.2025.11228391","relation":{},"subject":[],"published":{"date-parts":[[2025,6,30]]}}}