{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/expansionnet-v2-block-static-expansion-in","title":"Exploiting Multiple Sequence Lengths in Fast End to End Training for Image Captioning","arxiv_id":"2208.06551","date":"2022-08-13","proceeding":null,"authors":["Jia Cheng Hu","Roberto Cavicchioli","Alessandro Capotondi"],"abstract":"We introduce a method called the Expansion mechanism that processes the input unconstrained by the number of elements in the sequence. By doing so, the model can learn more effectively compared to traditional attention-based approaches. To support this claim, we design a novel architecture ExpansionNet v2 that achieved strong results on the MS COCO 2014 Image Captioning challenge and the State of the Art in its respective category, with a score of 143.7 CIDErD in the offline test split, 140.8 CIDErD in the online evaluation server and 72.9 AllCIDEr on the nocaps validation set. Additionally, we introduce an End to End training algorithm up to 2.8 times faster than established alternatives. Source code available at: https://github.com/jchenghu/ExpansionNet_v2","url_abs":"https://arxiv.org/abs/2208.06551v4","url_pdf":"https://arxiv.org/pdf/2208.06551v4.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"expansionnet-v2-block-static-expansion-in","repo_url":"https://github.com/jchenghu/expansionnet_v2","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"image-captioning","task_name":"Image Captioning"}],"methods":[{"method_slug":"test","method_name":"Test"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-captioning-on-coco","task":"Image Captioning","dataset":"COCO (Common Objects in Context)","model":"ExpansionNet v2","rank_in_archive_order":1,"of":17,"metrics":{"CIDEr":"143.7"},"uses_additional_data":false},{"leaderboard":"/sota/image-captioning-on-coco-captions","task":"Image Captioning","dataset":"COCO Captions","model":"ExpansionNet v2 (No VL pretraining)","rank_in_archive_order":6,"of":41,"metrics":{"BLEU-1":"83.5","BLEU-4":"42.7","CIDER":"143.7","METEOR":"30.6","ROUGE-L":"61.1","SPICE":"24.7"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2208.06551","atlas_url":"https://app.syntology.ai/?focus=2208.06551","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}