{"url":"/sota/image-paragraph-captioning-on-image-paragraph","task":{"name":"Image Paragraph Captioning","url":"/task/image-paragraph-captioning","note":null},"dataset":{"name":"Image Paragraph Captioning","url":"/dataset/image-paragraph-captioning"},"category":"Graphs","categories":["Graphs","Reasoning"],"category_note":null,"description":"Image paragraph captioning involves generating a detailed, multi-sentence description of the content of an image.","description_from":"task","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","rank":"the archive's row order at snapshot; not re-ranked","rows_end_at":"2025-07-28","rows_withheld_as_spam":0,"metric_values":"the archive's strings, untouched"},"metrics":["BLEU-4","METEOR","CIDEr"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"BLEU-4":"higher","METEOR":null,"CIDEr":null}},"counts":{"rows":10,"rows_with_code":2,"rows_with_paper_page":10,"rows_dated":9,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"HSGED(SLL)","metrics":{"BLEU-4":"11.26","CIDEr":"36.02","METEOR":"18.33"},"uses_additional_data":false,"paper_date":"2020-10-12","paper":"/paper/hierarchical-scene-graph-encoder-decoder-for","paper_url":"https://dl.acm.org/doi/abs/10.1145/3394171.3413859","paper_title":"Hierarchical Scene Graph Encoder-Decoder for Image Paragraph Captioning","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":2,"model":"SCST training, w/ rep. penalty","metrics":{"BLEU-4":"10.58","CIDEr":"30.63","METEOR":"17.86"},"uses_additional_data":false,"paper_date":"2018-10-01","paper":"/paper/training-for-diversity-in-image-paragraph","paper_url":"https://aclanthology.org/D18-1084","paper_title":"Training for Diversity in Image Paragraph Captioning","code":"https://github.com/lukemelas/image-paragraph-captioning","n_code_links":1,"syntology":null},{"rank_in_archive_order":3,"model":"IMAP","metrics":{"BLEU-4":"10.29","CIDEr":"24.07","METEOR":"17.36"},"uses_additional_data":false,"paper_date":"2020-12-01","paper":"/paper/interactive-key-value-memory-augmented","paper_url":"https://aclanthology.org/2020.coling-main.279","paper_title":"Interactive Key-Value Memory-augmented Attention for Image Paragraph Captioning","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":4,"model":"CAE-LSTM","metrics":{"BLEU-4":"9.67","CIDEr":"25.15","METEOR":"18.82"},"uses_additional_data":false,"paper_date":"2019-08-01","paper":"/paper/convolutional-auto-encoding-of-sentence","paper_url":"https://arxiv.org/abs/1908.00249v1","paper_title":"Convolutional Auto-encoding of Sentence Topics for Image Paragraph Generation","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":5,"model":"Diverse and Coherent Paragraph Generation from Images","metrics":{"BLEU-4":"9.43","CIDEr":"20.93","METEOR":"18.62"},"uses_additional_data":false,"paper_date":"2018-09-03","paper":"/paper/diverse-and-coherent-paragraph-generation","paper_url":"http://arxiv.org/abs/1809.00681v1","paper_title":"Diverse and Coherent Paragraph Generation from Images","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":6,"model":"RTT-GAN (Semi + Fully)","metrics":{"BLEU-4":"9.21","CIDEr":"20.36","METEOR":"18.39"},"uses_additional_data":false,"paper_date":"2017-03-21","paper":"/paper/recurrent-topic-transition-gan-for-visual","paper_url":"http://arxiv.org/abs/1703.07022v2","paper_title":"Recurrent Topic-Transition GAN for Visual Paragraph Generation","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":7,"model":"Regions-Hierarchical (ours)","metrics":{"BLEU-4":"8.69","CIDEr":"13.52","METEOR":"15.95"},"uses_additional_data":false,"paper_date":"2016-11-20","paper":"/paper/a-hierarchical-approach-for-generating","paper_url":"http://arxiv.org/abs/1611.06607v2","paper_title":"A Hierarchical Approach for Generating Descriptive Image Paragraphs","code":"https://github.com/chenxinpeng/im2p","n_code_links":3,"syntology":{"n_ran":0,"n_unverified":9,"n_samples":9,"n_pointer_only_licence":0}},{"rank_in_archive_order":8,"model":"Dual-CNN","metrics":{"BLEU-4":"8.6","CIDEr":"17.4","METEOR":"15.8"},"uses_additional_data":false,"paper_date":"2020-02-14","paper":"/paper/dual-cnn-a-convolutional-language-decoder-for","paper_url":"https://www.sciencedirect.com/science/article/pii/S0925231220302319","paper_title":"Dual-CNN: A Convolutional language decoder for paragraph image captioning","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":9,"model":"Depth-aware Attention Model (DAM)","metrics":{"BLEU-4":"6.7","CIDEr":"17.3","METEOR":"13.9"},"uses_additional_data":false,"paper_date":"2018-10-15","paper":"/paper/look-deeper-see-richer-depth-aware-image","paper_url":"https://dl.acm.org/doi/abs/10.1145/3240508.3240583","paper_title":"Look Deeper See Richer: Depth-aware Image Paragraph Captioning","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":10,"model":"IMG+LNG","metrics":{"BLEU-4":"4.67","CIDEr":"26.38","METEOR":"11.30"},"uses_additional_data":false,"paper_date":null,"paper":"/paper/when-an-image-tells-a-story-the-role-of","paper_url":"https://aclanthology.org/2020.inlg-1.40","paper_title":"When an Image Tells a Story: The Role of Visual and Semantic Information for Generating Paragraph Descriptions","code":null,"n_code_links":0,"syntology":null}],"since_archive":{"claim":"Results that newer papers report for their own method, placed here by Syntology. A model pointed at the cell in the paper's own table; the number was read from that cell and checked against this leaderboard's metric, dataset, split and scale; an independent check that saw this leaderboard's other rows and every other leaderboard on the same dataset accepted it. Not reviewed by the paper's authors or by the archive's editors, and not ranked against the archive rows.","extraction_file_present":true,"measurement":{"test_papers":883,"papers_with_output":881,"judged_true":108,"judged":110,"wilson95_lower":0.9361,"measured_on":"2026-09-24","frozen_commit":"0e3de0df94"},"measurement_note":"blind adjudication of accepted entries on a held-out split of archive papers, rules frozen before the test","coverage":{"sentence":"Syntology has checked 6,795 of the 9,581 papers on this site that are newer than the archive; results from the others appear after they are checked.","complete":false,"papers_newer_than_archive":9581,"papers_checked":6795,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":2785},"order":"newest first by month (arXiv date, else the arXiv-id month), then arXiv id descending","columns":[],"entries":[]},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per row: N of M harvested code samples from that row's paper executed on a synthesized fixture; the other M-N are unverified. Not a reproduction of the row's number; not a correctness claim. n_pointer_only_licence counts samples the site points at rather than redistributes (a licence axis, independent of ran/unverified).","rows_with_graph_line":1,"rows_with_any_sample_ran":0,"distinct_papers_with_graph_line":1,"distinct_papers_with_any_sample_ran":0,"samples_over_distinct_papers":{"n_ran":0,"n_unverified":9,"n_samples":9,"n_pointer_only_licence":0,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":0,"n_unverified":9,"n_samples":9,"n_pointer_only_licence":0,"note":"row-weighted: a paper behind several rows is counted once per row; inflated relative to samples_over_distinct_papers by design, kept for readers summing the per-row syntology blocks"}}}