{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,4]],"date-time":"2026-06-04T16:09:02Z","timestamp":1780589342309,"version":"3.54.1"},"reference-count":118,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"4","license":[{"start":{"date-parts":[[2025,4,1]],"date-time":"2025-04-01T00:00:00Z","timestamp":1743465600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2025,4,1]],"date-time":"2025-04-01T00:00:00Z","timestamp":1743465600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,4,1]],"date-time":"2025-04-01T00:00:00Z","timestamp":1743465600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100012166","name":"National Key Research and Development Program of China","doi-asserted-by":"publisher","award":["2023YFB4502804"],"award-info":[{"award-number":["2023YFB4502804"]}],"id":[{"id":"10.13039\/501100012166","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100014219","name":"National Science Fund for Distinguished Young Scholars","doi-asserted-by":"publisher","award":["62025603"],"award-info":[{"award-number":["62025603"]}],"id":[{"id":"10.13039\/501100014219","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["U21B2037"],"award-info":[{"award-number":["U21B2037"]}],"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":["U22B2051"],"award-info":[{"award-number":["U22B2051"]}],"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":["62072389"],"award-info":[{"award-number":["62072389"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"name":"National Natural Science Fund for Young Scholars of China","award":["62302411"],"award-info":[{"award-number":["62302411"]}]},{"DOI":"10.13039\/501100002858","name":"China Postdoctoral Science Foundation","doi-asserted-by":"publisher","award":["2023M732948"],"award-info":[{"award-number":["2023M732948"]}],"id":[{"id":"10.13039\/501100002858","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Natural Science Foundation of Fujian Province of China","award":["2021J01002"],"award-info":[{"award-number":["2021J01002"]}]},{"name":"Natural Science Foundation of Fujian Province of China","award":["2022J06001"],"award-info":[{"award-number":["2022J06001"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Neural Netw. Learning Syst."],"published-print":{"date-parts":[[2025,4]]},"DOI":"10.1109\/tnnls.2024.3409354","type":"journal-article","created":{"date-parts":[[2024,7,31]],"date-time":"2024-07-31T16:50:54Z","timestamp":1722444654000},"page":"6203-6217","source":"Crossref","is-referenced-by-count":12,"title":["Image Captioning via Dynamic Path Customization"],"prefix":"10.1109","volume":"36","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-8744-3423","authenticated-orcid":false,"given":"Yiwei","family":"Ma","sequence":"first","affiliation":[{"name":"Key Laboratory of Multimedia Trusted Perception and Efficient Computing, Ministry of Education of China, Xiamen University, Xiamen, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9956-6308","authenticated-orcid":false,"given":"Jiayi","family":"Ji","sequence":"additional","affiliation":[{"name":"Key Laboratory of Multimedia Trusted Perception and Efficient Computing, Ministry of Education of China, Xiamen University, Xiamen, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3912-9306","authenticated-orcid":false,"given":"Xiaoshuai","family":"Sun","sequence":"additional","affiliation":[{"name":"Key Laboratory of Multimedia Trusted Perception and Efficient Computing, Ministry of Education of China, Xiamen University, Xiamen, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5110-4526","authenticated-orcid":false,"given":"Yiyi","family":"Zhou","sequence":"additional","affiliation":[{"name":"Key Laboratory of Multimedia Trusted Perception and Efficient Computing, Ministry of Education of China, Xiamen University, Xiamen, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0611-0636","authenticated-orcid":false,"given":"Xiaopeng","family":"Hong","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology, Harbin Institute of Technology, Harbin, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yongjian","family":"Wu","sequence":"additional","affiliation":[{"name":"Youtu Laboratory, Tencent, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9163-2932","authenticated-orcid":false,"given":"Rongrong","family":"Ji","sequence":"additional","affiliation":[{"name":"Key Laboratory of Multimedia Trusted Perception and Efficient Computing, Ministry of Education of China, Xiamen University, Xiamen, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2018.2855406"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2018.2889922"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2020.2969330"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2019.2909864"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2018.2824816"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2020.3042192"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/TMM.2020.3011317"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2023.109420"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/TMM.2019.2896516"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/TMM.2018.2869276"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2022.3183434"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/tmm.2022.3164787"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/tnnls.2022.3185320"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/tnnls.2022.3152990"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2020.2984972"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00728"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58452-8_22"},{"key":"ref18","article-title":"CondConv: Conditionally parameterized convolutions for efficient inference","author":"Yang","year":"2019","journal-title":"arXiv:1904.04971"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01104"},{"key":"ref20","article-title":"DyNet: Dynamic convolution for accelerating convolutional neural networks","author":"Zhang","year":"2020","journal-title":"arXiv:2004.10694"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00797"},{"key":"ref22","article-title":"Revisiting dynamic convolution via matrix decomposition","author":"Li","year":"2021","journal-title":"arXiv:2103.08756"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00208"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00745"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1613\/jair.3994"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00166"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01028"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.345"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01264-9_42"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01216-8_31"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v32i1.12266"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v33i01.33018142"},{"key":"ref33","article-title":"Image captioning: Transforming objects into words","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"32","author":"Herdade"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.01094"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01098"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v35i3.16328"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01521"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i07.7005"},{"key":"ref39","article-title":"SimVLM: Simple visual language model pretraining with weak supervision","author":"Wang","year":"2021","journal-title":"arXiv:2108.10904"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00553"},{"key":"ref41","article-title":"M6: A Chinese multimodal pretrainer","author":"Lin","year":"2021","journal-title":"arXiv:2103.00823"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1145\/3503161.3547910"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.131"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00636"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00473"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01059"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58577-8_8"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D13-1128"},{"key":"ref49","first-page":"747","article-title":"Midge: Generating image descriptions from computer vision detections","volume-title":"Proc. 13th Conf. Eur. Chapter Assoc. Comput. Linguistics","author":"Mitchell"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1145\/3581783.3611768"},{"key":"ref51","article-title":"Exploring nearest neighbor approaches for image captioning","author":"Devlin","year":"2015","journal-title":"arXiv:1505.04467"},{"key":"ref52","first-page":"1889","article-title":"Deep fragment embeddings for bidirectional image sentence mapping","volume-title":"Proc. 27th Int. Conf. Neural Inf. Process. Syst.","author":"Karpathy"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2015.7298935"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.1015666108"},{"key":"ref55","doi-asserted-by":"crossref","first-page":"127","DOI":"10.1016\/j.conb.2020.03.004","article-title":"Parallel processing streams in the hippocampus","volume":"64","author":"Lee","year":"2020","journal-title":"Current Opinion Neurobiol."},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00244"},{"key":"ref57","article-title":"Multi-scale dense networks for resource efficient image classification","author":"Huang","year":"2017","journal-title":"arXiv:1703.09844"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1145\/3581783.3611767"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1117\/12.3022812"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1145\/3581783.3612259"},{"key":"ref61","article-title":"Rotated multi-scale interaction network for referring remote sensing image segmentation","author":"Liu","year":"2023","journal-title":"arXiv:2312.12470"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00858"},{"key":"ref63","article-title":"3D-STMN: Dependency-driven superpoint-text matching network for end-to-end 3D referring expression segmentation","author":"Wu","year":"2023","journal-title":"arXiv:2308.16632"},{"key":"ref64","article-title":"ELF: An early-exiting framework for long-tailed classification","author":"Duggal","year":"2020","journal-title":"arXiv:2006.11979"},{"key":"ref65","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.1706.03762"},{"key":"ref66","first-page":"22682","article-title":"Aligning pretraining for detection via object-level contrastive learning","volume":"34","author":"Wei","year":"2021","journal-title":"Proc. Adv. Neural Inf. Process. Syst."},{"key":"ref67","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00986"},{"key":"ref68","article-title":"CoAtNet: Marrying convolution and attention for all data sizes","author":"Dai","year":"2021","journal-title":"arXiv:2106.04803"},{"key":"ref69","article-title":"CSWin transformer: A general vision transformer backbone with cross-shaped windows","author":"Dong","year":"2021","journal-title":"arXiv:2107.00652"},{"key":"ref70","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00042"},{"key":"ref71","article-title":"MobileViT: Light-weight, general-purpose, and mobile-friendly vision transformer","author":"Mehta","year":"2021","journal-title":"arXiv:2110.02178"},{"key":"ref72","doi-asserted-by":"publisher","DOI":"10.5555\/3045118.3045167"},{"key":"ref73","article-title":"Sparse MLP for image recognition: Is self-attention really necessary?","author":"Tang","year":"2021","journal-title":"arXiv:2109.05422"},{"key":"ref74","doi-asserted-by":"publisher","DOI":"10.5555\/3104322.3104425"},{"key":"ref75","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2015.7299087"},{"key":"ref76","first-page":"311","article-title":"BLEU: A method for automatic evaluation of machine translation","volume-title":"Proc. 40th Annu. Meeting Assoc. Comput. Linguistics","author":"Papineni"},{"key":"ref77","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-10602-1_48"},{"key":"ref78","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2015.7298932"},{"key":"ref79","first-page":"91","article-title":"Faster R-CNN: Towards real-time object detection with region proposal networks","volume-title":"Proc. Int. Conf. Adv. Neural Inf. Process. Syst.","volume":"28","author":"Ren"},{"key":"ref80","article-title":"Adam: A method for stochastic optimization","author":"Kingma","year":"2014","journal-title":"arXiv:1412.6980"},{"key":"ref81","first-page":"65","article-title":"METEOR: An automatic metric for MT evaluation with improved correlation with human judgments","volume-title":"Proc. ACL Workshop Intrinsic Extrinsic Eval. Measures Mach. Transl. Summarization","author":"Banerjee"},{"key":"ref82","first-page":"74","article-title":"Rouge: A package for automatic evaluation of summaries","volume-title":"Proc. Workshop Text Summarization ACL","author":"Lin"},{"key":"ref83","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46454-1_24"},{"key":"ref84","article-title":"How much can CLIP benefit vision-and-language tasks?","author":"Shen","year":"2021","journal-title":"arXiv:2107.06383"},{"key":"ref85","first-page":"12888","article-title":"BLIP: Bootstrapping language-image pre-training for unified vision-language understanding and generation","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Li"},{"key":"ref86","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01745"},{"key":"ref87","first-page":"23318","article-title":"OFA: Unifying architectures, tasks, and modalities through a simple sequence-to-sequence learning framework","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Wang"},{"key":"ref88","doi-asserted-by":"publisher","DOI":"10.24963\/ijcai.2020\/107"},{"key":"ref89","doi-asserted-by":"publisher","DOI":"10.1109\/ICCVW54120.2021.00350"},{"key":"ref90","doi-asserted-by":"publisher","DOI":"10.24963\/ijcai.2021\/91"},{"key":"ref91","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01190"},{"key":"ref92","doi-asserted-by":"publisher","DOI":"10.1109\/tmm.2022.3154149"},{"key":"ref93","doi-asserted-by":"publisher","DOI":"10.24963\/ijcai.2022\/224"},{"key":"ref94","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v37i1.25137"},{"key":"ref95","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.02237"},{"key":"ref96","doi-asserted-by":"publisher","DOI":"10.1109\/TMM.2023.3295098"},{"key":"ref97","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.524"},{"key":"ref98","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00902"},{"key":"ref99","article-title":"Categorical reparameterization with gumbel-softmax","author":"Jang","year":"2016","journal-title":"arXiv:1611.01144"},{"key":"ref100","doi-asserted-by":"publisher","DOI":"10.5555\/3045118.3045336"},{"key":"ref101","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.277"},{"key":"ref102","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2015.7298878"},{"key":"ref103","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.634"},{"key":"ref104","article-title":"Deep captioning with multimodal recurrent neural networks (m-RNN)","author":"Mao","year":"2014","journal-title":"arXiv:1412.6632"},{"key":"ref105","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.503"},{"key":"ref106","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00444"},{"key":"ref107","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.202"},{"key":"ref108","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2018.2817340"},{"key":"ref109","article-title":"Bilinear attention networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"31","author":"Kim"},{"key":"ref110","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00644"},{"key":"ref111","first-page":"1931","article-title":"Unifying vision-and-language tasks via text generation","volume-title":"Proc. 38th Int. Conf. Mach. Learn.","volume":"139","author":"Cho"},{"key":"ref112","first-page":"200","article-title":"Multimodal few-shot learning with frozen language models","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Tsimpoukelli"},{"key":"ref113","article-title":"Language models are general-purpose interfaces","author":"Hao","year":"2022","journal-title":"arXiv:2206.06336"},{"key":"ref114","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2022.findings-acl.187"},{"key":"ref115","article-title":"A good prompt is worth millions of parameters: Low-resource prompt-based learning for vision-language models","author":"Jin","year":"2021","journal-title":"arXiv:2110.08484"},{"key":"ref116","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2022.findings-emnlp.67"},{"key":"ref117","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01046"},{"key":"ref118","article-title":"BLIP-2: Bootstrapping language-image pre-training with frozen image encoders and large language models","author":"Li","year":"2023","journal-title":"arXiv:2301.12597"}],"container-title":["IEEE Transactions on Neural Networks and Learning Systems"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/5962385\/10949581\/10616183.pdf?arnumber=10616183","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,5]],"date-time":"2025-12-05T18:39:19Z","timestamp":1764959959000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10616183\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,4]]},"references-count":118,"journal-issue":{"issue":"4"},"URL":"https:\/\/doi.org\/10.1109\/tnnls.2024.3409354","relation":{},"ISSN":["2162-237X","2162-2388"],"issn-type":[{"value":"2162-237X","type":"print"},{"value":"2162-2388","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,4]]}}}