{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,4]],"date-time":"2026-08-04T07:48:50Z","timestamp":1785829730372,"version":"3.56.0"},"reference-count":50,"publisher":"Elsevier BV","license":[{"start":{"date-parts":[[2026,12,1]],"date-time":"2026-12-01T00:00:00Z","timestamp":1796083200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/tdm\/userlicense\/1.0\/"},{"start":{"date-parts":[[2026,12,1]],"date-time":"2026-12-01T00:00:00Z","timestamp":1796083200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/legal\/tdmrep-license"},{"start":{"date-parts":[[2026,12,1]],"date-time":"2026-12-01T00:00:00Z","timestamp":1796083200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-017"},{"start":{"date-parts":[[2026,12,1]],"date-time":"2026-12-01T00:00:00Z","timestamp":1796083200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"},{"start":{"date-parts":[[2026,12,1]],"date-time":"2026-12-01T00:00:00Z","timestamp":1796083200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-012"},{"start":{"date-parts":[[2026,12,1]],"date-time":"2026-12-01T00:00:00Z","timestamp":1796083200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,12,1]],"date-time":"2026-12-01T00:00:00Z","timestamp":1796083200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-004"}],"funder":[{"DOI":"10.13039\/501100001459","name":"Ministry of Education - Singapore","doi-asserted-by":"publisher","award":["MOE-T2EP20123-0005"],"award-info":[{"award-number":["MOE-T2EP20123-0005"]}],"id":[{"id":"10.13039\/501100001459","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001348","name":"Agency for Science, Technology and Research","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100001348","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001475","name":"Nanyang Technological University","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100001475","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["elsevier.com","sciencedirect.com"],"crossmark-restriction":true},"short-container-title":["Information Fusion"],"published-print":{"date-parts":[[2026,12]]},"DOI":"10.1016\/j.inffus.2026.104550","type":"journal-article","created":{"date-parts":[[2026,6,10]],"date-time":"2026-06-10T07:35:05Z","timestamp":1781076905000},"page":"104550","update-policy":"https:\/\/doi.org\/10.1016\/elsevier_cm_policy","source":"Crossref","is-referenced-by-count":0,"special_numbering":"C","title":["Multi-agent SEA: step-wise evidence acquisition for multimodal financial reasoning"],"prefix":"10.1016","volume":"136","author":[{"ORCID":"https:\/\/orcid.org\/0009-0004-4395-2877","authenticated-orcid":false,"given":"Shuangyan","family":"Deng","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2419-3332","authenticated-orcid":false,"given":"Zihao","family":"Huang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7856-3140","authenticated-orcid":false,"given":"Kelvin","family":"Du","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1082-8755","authenticated-orcid":false,"given":"Rui","family":"Mao","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0824-0899","authenticated-orcid":false,"given":"Jiamou","family":"Liu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5512-0868","authenticated-orcid":false,"given":"Ciprian Doru","family":"Giurc\u0103neanu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Erik","family":"Cambria","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"78","reference":[{"key":"10.1016\/j.inffus.2026.104550_sbref0001","series-title":"GPT-4o system card","author":"AI","year":"2024"},{"key":"10.1016\/j.inffus.2026.104550_bib0002","unstructured":"G. DeepMind, Gemini 2.5 flash model card, 2025, https:\/\/deepmind.google\/models\/model-cards\/."},{"key":"10.1016\/j.inffus.2026.104550_bib0003","unstructured":"Anthro pic, Claude Sonnet 4-5, 2025, https:\/\/www.anthropic.com\/news\/claude-4."},{"key":"10.1016\/j.inffus.2026.104550_bib0004","doi-asserted-by":"crossref","DOI":"10.1016\/j.inffus.2024.102755","article-title":"Natural language processing in finance: a survey","volume":"115","author":"Du","year":"2025","journal-title":"Inform. Fus."},{"key":"10.1016\/j.inffus.2026.104550_bib0005","doi-asserted-by":"crossref","first-page":"34892","DOI":"10.52202\/075280-1516","article-title":"Visual instruction tuning","volume":"36","author":"Liu","year":"2023","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"10.1016\/j.inffus.2026.104550_bib0006","article-title":"InterARM: interpretable affective reasoning model for multimodal sarcasm detection","author":"Yue","year":"2026","journal-title":"IEEE Trans. Affective Comput."},{"key":"10.1016\/j.inffus.2026.104550_sbref0007","series-title":"FAMMA: a benchmark for financial domain multilingual multimodal question answering","author":"Xue","year":"2024"},{"key":"10.1016\/j.inffus.2026.104550_bib0008","series-title":"Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics","first-page":"11640","article-title":"FinMME: benchmark dataset for financial multi-modal reasoning evaluation","author":"Luo","year":"2025"},{"key":"10.1016\/j.inffus.2026.104550_bib0009","series-title":"Proceedings of the 6th ACM International Conference on AI in Finance","first-page":"168","article-title":"FinMR: a knowledge-intensive multimodal benchmark for advanced financial reasoning","author":"Deng","year":"2025"},{"key":"10.1016\/j.inffus.2026.104550_bib0010","article-title":"FinQA: a dataset of numerical reasoning over financial data","author":"Chen","year":"2021","journal-title":"Proceedings EMNLP 2021"},{"key":"10.1016\/j.inffus.2026.104550_bib0011","series-title":"Proceedings of the 17th ACM International Conference on Web Search and Data Mining","first-page":"645","article-title":"Table meets LLM: can large language models understand structured table data? A benchmark and empirical study","author":"Sui","year":"2024"},{"key":"10.1016\/j.inffus.2026.104550_sbref0012","series-title":"Proceedings of the IEEE\/CVF International Conference on Computer Vision","first-page":"3245","article-title":"FinMMR: make financial numerical reasoning more multimodal, comprehensive, and challenging","author":"Tang","year":"2025"},{"issue":"6","key":"10.1016\/j.inffus.2026.104550_bib0013","doi-asserted-by":"crossref","first-page":"62","DOI":"10.1109\/MIS.2023.3329745","article-title":"Seven pillars for the future of artificial intelligence","volume":"38","author":"Cambria","year":"2023","journal-title":"IEEE Intell. Syst."},{"key":"10.1016\/j.inffus.2026.104550_bib0014","series-title":"Psychology of Learning and Motivation","doi-asserted-by":"crossref","first-page":"89","DOI":"10.1016\/S0079-7421(08)60422-3","article-title":"Human memory: a proposed system and its control processes","volume":"2","author":"Atkinson","year":"1968"},{"key":"10.1016\/j.inffus.2026.104550_bib0015","series-title":"Findings of the Association for Computational Linguistics: aCL, Bangkok, Thailand","first-page":"9891","article-title":"MetaPro 2.0: computational metaphor processing on the effectiveness of anomalous language modeling","author":"Mao","year":"2024"},{"issue":"3","key":"10.1016\/j.inffus.2026.104550_bib0016","doi-asserted-by":"crossref","first-page":"201","DOI":"10.1038\/nrn755","article-title":"Control of goal-directed and stimulus-driven attention in the brain","volume":"3","author":"Corbetta","year":"2002","journal-title":"Nat. Rev. Neurosci."},{"key":"10.1016\/j.inffus.2026.104550_sbref0017","series-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","first-page":"9556","article-title":"MMMU: a massive multi-discipline multimodal understanding and reasoning benchmark for expert AGI","author":"Yue","year":"2024"},{"key":"10.1016\/j.inffus.2026.104550_bib0018","series-title":"Proceedings of the 6th ACM International Conference on AI in Finance","first-page":"483","article-title":"A role-aware multi-agent framework for financial education QA","author":"Zhu","year":"2025"},{"key":"10.1016\/j.inffus.2026.104550_bib0019","series-title":"First Conference on Language Modeling","first-page":"1","article-title":"AutoGen: enabling next-gen LLM applications via multi-agent conversations","author":"Wu","year":"2024"},{"key":"10.1016\/j.inffus.2026.104550_bib0020","series-title":"Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing","first-page":"17889","article-title":"Encouraging divergent thinking in large language models through multi-agent debate","author":"Liang","year":"2024"},{"key":"10.1016\/j.inffus.2026.104550_sbref0021","series-title":"BloombergGPT: a large language model for finance","author":"Wu","year":"2023"},{"key":"10.1016\/j.inffus.2026.104550_bib0022","series-title":"Proceedings of the 33rd ACM International Conference on Information and Knowledge Management (CIKM), Idaho, USA","first-page":"529","article-title":"Explainable stock price movement prediction using contrastive learning","author":"Du","year":"2024"},{"issue":"3","key":"10.1016\/j.inffus.2026.104550_bib0023","doi-asserted-by":"crossref","first-page":"4086","DOI":"10.1109\/TCSS.2026.3677113","article-title":"SenticNet 9: generative commonsense for emotion AI via conceptual primitive discovery and time shift mechanism","volume":"13","author":"Cambria","year":"2026","journal-title":"IEEE Trans. Comput. Soc. Syst."},{"key":"10.1016\/j.inffus.2026.104550_bib0024","series-title":"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: long Papers)","first-page":"3277","article-title":"TAT-QA: a question answering benchmark on a hybrid of tabular and textual content in finance","author":"Zhu","year":"2021"},{"key":"10.1016\/j.inffus.2026.104550_bib0025","doi-asserted-by":"crossref","first-page":"15","DOI":"10.1109\/MIS.2025.3544912","article-title":"A retrieval-augmented multi-agent system for financial sentiment analysis","volume":"40","author":"Du","year":"2025","journal-title":"IEEE Intell. Syst."},{"key":"10.1016\/j.inffus.2026.104550_bib0026","doi-asserted-by":"crossref","DOI":"10.1109\/TAFFC.2025.3565506","article-title":"Exploring cognitive and aesthetic causality for multimodal aspect-based sentiment analysis","author":"Xiao","year":"2025","journal-title":"IEEE Trans. Affective Comput."},{"key":"10.1016\/j.inffus.2026.104550_sbref0027","series-title":"TradingAgents: multi-agents LLM financial trading framework","author":"Xiao","year":"2024"},{"key":"10.1016\/j.inffus.2026.104550_sbref0028","series-title":"Large language model agent in financial trading: a survey","author":"Ding","year":"2024"},{"key":"10.1016\/j.inffus.2026.104550_sbref0029","series-title":"Adaptive computation time for recurrent neural networks","author":"Graves","year":"2016"},{"key":"10.1016\/j.inffus.2026.104550_bib0030","series-title":"Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics","first-page":"6640","article-title":"The Right Tool for the Job: matching model and instance complexities","author":"Schwartz","year":"2020"},{"key":"10.1016\/j.inffus.2026.104550_bib0031","series-title":"Proceedings of the AAAI Conference on Artificial Intelligence","first-page":"3959","article-title":"Classification with costly features using deep reinforcement learning","volume":"33","author":"Janisch","year":"2019"},{"key":"10.1016\/j.inffus.2026.104550_sbref0032","series-title":"Dropout feature ranking for deep learning models","author":"Chang","year":"2017"},{"key":"10.1016\/j.inffus.2026.104550_bib0033","article-title":"Recurrent models of visual attention","volume":"27","author":"Mnih","year":"2014","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"10.1016\/j.inffus.2026.104550_sbref0034","series-title":"Multiple object recognition with visual attention","author":"Ba","year":"2015"},{"key":"10.1016\/j.inffus.2026.104550_bib0035","series-title":"European Conference on Computer Vision","first-page":"396","article-title":"Adaptive token sampling for efficient vision transformers","author":"Fayyaz","year":"2022"},{"key":"10.1016\/j.inffus.2026.104550_bib0036","series-title":"Findings of the Association for Computational Linguistics: aCL 2024","first-page":"15890","article-title":"Visual in-context learning for large vision-language models","author":"Zhou","year":"2024"},{"key":"10.1016\/j.inffus.2026.104550_bib0037","series-title":"The Eleventh International Conference on Learning Representations","first-page":"11640","article-title":"React: synergizing reasoning and acting in language models","author":"Yao","year":"2022"},{"key":"10.1016\/j.inffus.2026.104550_bib0038","doi-asserted-by":"crossref","first-page":"68539","DOI":"10.52202\/075280-2997","article-title":"Toolformer: language models can teach themselves to use tools","volume":"36","author":"Schick","year":"2023","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"10.1016\/j.inffus.2026.104550_bib0039","series-title":"Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: long Papers)","first-page":"18305","article-title":"RARE: retrieval-augmented reasoning enhancement for large language models","author":"Tran","year":"2025"},{"key":"10.1016\/j.inffus.2026.104550_bib0040","doi-asserted-by":"crossref","DOI":"10.1016\/j.knosys.2022.109494","article-title":"Entity alignment based on relational semantics augmentation for multilingual knowledge graphs","volume":"252","author":"Akhtar","year":"2022","journal-title":"Knowl.-Based Syst."},{"key":"10.1016\/j.inffus.2026.104550_bib0041","doi-asserted-by":"crossref","DOI":"10.1016\/j.engappai.2024.109660","article-title":"Multilingual entity alignment by abductive knowledge reasoning on multiple knowledge graphs","volume":"139","author":"Akhtar","year":"2025","journal-title":"Eng. Appl. Artif. Intell."},{"key":"10.1016\/j.inffus.2026.104550_bib0042","series-title":"Proceedings of the 36th Annual ACM Symposium on User Interface Software and Technology","first-page":"1","article-title":"Generative agents: interactive simulacra of human behavior","author":"Park","year":"2023"},{"key":"10.1016\/j.inffus.2026.104550_bib0043","series-title":"Findings of the Association for Computational Linguistics: aCL 2025","first-page":"25319","article-title":"MAM: modular multi-agent framework for multi-modal medical diagnosis via role-specialized collaboration","author":"Zhou","year":"2025"},{"key":"10.1016\/j.inffus.2026.104550_bib0044","doi-asserted-by":"crossref","first-page":"24824","DOI":"10.52202\/068431-1800","article-title":"Chain-of-thought prompting elicits reasoning in large language models","volume":"35","author":"Wei","year":"2022","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"10.1016\/j.inffus.2026.104550_bib0045","series-title":"Findings of the Association for Computational Linguistics: aCL 2025","first-page":"11640","article-title":"Voting or consensus? decision-making in multi-agent debate","author":"Kaesberg","year":"2025"},{"key":"10.1016\/j.inffus.2026.104550_bib0046","series-title":"Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing","first-page":"2511","article-title":"G-eval: NLG evaluation using GPT-4 with better human alignment","author":"Liu","year":"2023"},{"key":"10.1016\/j.inffus.2026.104550_bib0047","series-title":"Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (ACL)","first-page":"14717","article-title":"QAEval: mixture of evaluators for question-answering task evaluation","author":"Yue","year":"2025"},{"key":"10.1016\/j.inffus.2026.104550_bib0048","series-title":"The Eleventh International Conference on Learning Representations","first-page":"1","article-title":"Least-to-Most prompting enables complex reasoning in large language models","author":"Zhou","year":"2023"},{"key":"10.1016\/j.inffus.2026.104550_bib0049","series-title":"Findings of the Association for Computational Linguistics: eMNLP 2024","first-page":"13838","article-title":"Skills-in-context: unlocking compositionality in large language models","author":"Chen","year":"2024"},{"key":"10.1016\/j.inffus.2026.104550_bib0050","series-title":"The Thirty-eighth Annual Conference on Neural Information Processing Systems","first-page":"1","article-title":"DARG: dynamic evaluation of large language models via adaptive reasoning graph","author":"Zhang","year":"2024"}],"container-title":["Information Fusion"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S1566253526004288?httpAccept=text\/xml","content-type":"text\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S1566253526004288?httpAccept=text\/plain","content-type":"text\/plain","content-version":"vor","intended-application":"text-mining"}],"deposited":{"date-parts":[[2026,8,4]],"date-time":"2026-08-04T07:26:31Z","timestamp":1785828391000},"score":1,"resource":{"primary":{"URL":"https:\/\/linkinghub.elsevier.com\/retrieve\/pii\/S1566253526004288"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,12]]},"references-count":50,"alternative-id":["S1566253526004288"],"URL":"https:\/\/doi.org\/10.1016\/j.inffus.2026.104550","relation":{},"ISSN":["1566-2535"],"issn-type":[{"value":"1566-2535","type":"print"}],"subject":[],"published":{"date-parts":[[2026,12]]},"assertion":[{"value":"Elsevier","name":"publisher","label":"This article is maintained by"},{"value":"Multi-agent SEA: step-wise evidence acquisition for multimodal financial reasoning","name":"articletitle","label":"Article Title"},{"value":"Information Fusion","name":"journaltitle","label":"Journal Title"},{"value":"https:\/\/doi.org\/10.1016\/j.inffus.2026.104550","name":"articlelink","label":"CrossRef DOI link to publisher maintained version"},{"value":"article","name":"content_type","label":"Content Type"},{"value":"\u00a9 2026 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.","name":"copyright","label":"Copyright"}],"article-number":"104550"}}