{"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/training-for-diversity-in-image-paragraph","title":"Training for Diversity in Image Paragraph Captioning","arxiv_id":null,"date":"2018-10-01","proceeding":"EMNLP 2018 10","authors":["Luke Melas-Kyriazi","Alex Rush","er","George Han"],"abstract":"Image paragraph captioning models aim to produce detailed descriptions of a source image. These models use similar techniques as standard image captioning models, but they have encountered issues in text generation, notably a lack of diversity between sentences, that have limited their effectiveness. In this work, we consider applying sequence-level training for this task. We find that standard self-critical training produces poor results, but when combined with an integrated penalty on trigram repetition produces much more diverse paragraphs. This simple training approach improves on the best result on the Visual Genome paragraph captioning dataset from 16.9 to 30.6 CIDEr, with gains on METEOR and BLEU as well, without requiring any architectural changes.","url_abs":"https://aclanthology.org/D18-1084","url_pdf":"https://aclanthology.org/D18-1084.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":"training-for-diversity-in-image-paragraph","repo_url":"https://github.com/lukemelas/image-paragraph-captioning","is_official":0,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"diversity","task_name":"Diversity"},{"task_slug":"image-captioning","task_name":"Image Captioning"},{"task_slug":"image-paragraph-captioning","task_name":"Image Paragraph Captioning"},{"task_slug":"machine-translation","task_name":"Machine Translation"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"policy-gradient-methods","task_name":"Policy Gradient Methods"},{"task_slug":"text-generation","task_name":"Text Generation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-paragraph-captioning-on-image-paragraph","task":"Image Paragraph Captioning","dataset":"Image Paragraph Captioning","model":"SCST training, w/ rep. penalty","rank_in_archive_order":2,"of":10,"metrics":{"BLEU-4":"10.58","CIDEr":"30.63","METEOR":"17.86"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}