{"url":"/sota/chart-question-answering-on-chartqa","task":{"name":"Chart Question Answering","url":"/task/chart-question-answering","note":null},"dataset":{"name":"ChartQA","url":"/dataset/chartqa"},"category":"Computer Code","categories":["Computer Code"],"category_note":null,"description":"Question Answering task on charts images","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":["1:1 Accuracy"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"1:1 Accuracy":"higher"}},"counts":{"rows":27,"rows_with_code":22,"rows_with_paper_page":27,"rows_dated":27,"rows_using_additional_data":13},"rows":[{"rank_in_archive_order":1,"model":"ChartPaLI-5B + PaLM 2-S","metrics":{"1:1 Accuracy":"81.3"},"uses_additional_data":true,"paper_date":"2024-03-19","paper":"/paper/chart-based-reasoning-transferring","paper_url":"https://arxiv.org/abs/2403.12596v1","paper_title":"Chart-based Reasoning: Transferring Capabilities from LLMs to VLMs","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":2,"model":"Gemini Ultra","metrics":{"1:1 Accuracy":"80.8"},"uses_additional_data":false,"paper_date":"2023-12-19","paper":"/paper/gemini-a-family-of-highly-capable-multimodal-1","paper_url":"https://arxiv.org/abs/2312.11805v5","paper_title":"Gemini: A Family of Highly Capable Multimodal Models","code":"https://github.com/valdecy/pybibx","n_code_links":1,"syntology":null},{"rank_in_archive_order":3,"model":"DePlot+FlanPaLM+Codex (PoT Self-Consistency)","metrics":{"1:1 Accuracy":"79.3"},"uses_additional_data":false,"paper_date":"2022-12-20","paper":"/paper/deplot-one-shot-visual-language-reasoning-by","paper_url":"https://arxiv.org/abs/2212.10505v2","paper_title":"DePlot: One-shot visual language reasoning by plot-to-table translation","code":"https://github.com/huggingface/transformers","n_code_links":1,"syntology":null},{"rank_in_archive_order":4,"model":"ChartPaLI-5B","metrics":{"1:1 Accuracy":"77.3"},"uses_additional_data":true,"paper_date":"2024-03-19","paper":"/paper/chart-based-reasoning-transferring","paper_url":"https://arxiv.org/abs/2403.12596v1","paper_title":"Chart-based Reasoning: Transferring Capabilities from LLMs to VLMs","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":5,"model":"DePlot+Codex (PoT Self-Consistency)","metrics":{"1:1 Accuracy":"76.7"},"uses_additional_data":false,"paper_date":"2022-12-20","paper":"/paper/deplot-one-shot-visual-language-reasoning-by","paper_url":"https://arxiv.org/abs/2212.10505v2","paper_title":"DePlot: One-shot visual language reasoning by plot-to-table translation","code":"https://github.com/huggingface/transformers","n_code_links":1,"syntology":null},{"rank_in_archive_order":6,"model":"ScreenAI 5B (4.62 B params, w/ OCR)","metrics":{"1:1 Accuracy":"76.7"},"uses_additional_data":true,"paper_date":"2024-02-07","paper":"/paper/screenai-a-vision-language-model-for-ui-and","paper_url":"https://arxiv.org/abs/2402.04615v3","paper_title":"ScreenAI: A Vision-Language Model for UI and Infographics Understanding","code":"https://github.com/google-research-datasets/screen_qa","n_code_links":2,"syntology":null},{"rank_in_archive_order":7,"model":"SMoLA-PaLI-X Specialist Model","metrics":{"1:1 Accuracy":"74.6"},"uses_additional_data":true,"paper_date":"2023-12-01","paper":"/paper/omni-smola-boosting-generalist-multimodal","paper_url":"https://arxiv.org/abs/2312.00968v2","paper_title":"Omni-SMoLA: Boosting Generalist Multimodal Models with Soft Mixture of Low-rank Experts","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":8,"model":"SMoLA-PaLI-X Generalist Model","metrics":{"1:1 Accuracy":"73.8"},"uses_additional_data":true,"paper_date":"2023-12-01","paper":"/paper/omni-smola-boosting-generalist-multimodal","paper_url":"https://arxiv.org/abs/2312.00968v2","paper_title":"Omni-SMoLA: Boosting Generalist Multimodal Models with Soft Mixture of Low-rank Experts","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":9,"model":"MatCha4096 + LaMenDa","metrics":{"1:1 Accuracy":"72.64"},"uses_additional_data":true,"paper_date":"2024-01-01","paper":"/paper/synthesize-step-by-step-tools-templates-and-1","paper_url":"http://openaccess.thecvf.com//content/CVPR2024/html/Li_Synthesize_Step-by-Step_Tools_Templates_and_LLMs_as_Data_Generators_for_CVPR_2024_paper.html","paper_title":"Synthesize Step-by-Step: Tools Templates and LLMs as Data Generators for Reasoning-Based Chart VQA","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":10,"model":"PaLI-X (Single-task FT w/ OCR)","metrics":{"1:1 Accuracy":"72.3"},"uses_additional_data":true,"paper_date":"2023-05-29","paper":"/paper/pali-x-on-scaling-up-a-multilingual-vision","paper_url":"https://arxiv.org/abs/2305.18565v1","paper_title":"PaLI-X: On Scaling up a Multilingual Vision and Language Model","code":"https://github.com/kyegomez/PALI","n_code_links":2,"syntology":{"n_ran":6,"n_unverified":1,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":11,"model":"PaLI-X (Single-task FT)","metrics":{"1:1 Accuracy":"70.9"},"uses_additional_data":true,"paper_date":"2023-05-29","paper":"/paper/pali-x-on-scaling-up-a-multilingual-vision","paper_url":"https://arxiv.org/abs/2305.18565v1","paper_title":"PaLI-X: On Scaling up a Multilingual Vision and Language Model","code":"https://github.com/kyegomez/PALI","n_code_links":2,"syntology":{"n_ran":6,"n_unverified":1,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":12,"model":"PaLI-X (Multi-task FT)","metrics":{"1:1 Accuracy":"70.6"},"uses_additional_data":true,"paper_date":"2023-05-29","paper":"/paper/pali-x-on-scaling-up-a-multilingual-vision","paper_url":"https://arxiv.org/abs/2305.18565v1","paper_title":"PaLI-X: On Scaling up a Multilingual Vision and Language Model","code":"https://github.com/kyegomez/PALI","n_code_links":2,"syntology":{"n_ran":6,"n_unverified":1,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":13,"model":"DePlot+FlanPaLM (Self-Consistency)","metrics":{"1:1 Accuracy":"70.5"},"uses_additional_data":false,"paper_date":"2022-12-20","paper":"/paper/deplot-one-shot-visual-language-reasoning-by","paper_url":"https://arxiv.org/abs/2212.10505v2","paper_title":"DePlot: One-shot visual language reasoning by plot-to-table translation","code":"https://github.com/huggingface/transformers","n_code_links":1,"syntology":null},{"rank_in_archive_order":14,"model":"PaLI-3","metrics":{"1:1 Accuracy":"70"},"uses_additional_data":false,"paper_date":"2023-10-13","paper":"/paper/pali-3-vision-language-models-smaller-faster","paper_url":"https://arxiv.org/abs/2310.09199v2","paper_title":"PaLI-3 Vision Language Models: Smaller, Faster, Stronger","code":"https://github.com/kyegomez/PALI3","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":0,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":15,"model":"PaLI-3 (w/ OCR)","metrics":{"1:1 Accuracy":"69.5"},"uses_additional_data":false,"paper_date":"2023-10-13","paper":"/paper/pali-3-vision-language-models-smaller-faster","paper_url":"https://arxiv.org/abs/2310.09199v2","paper_title":"PaLI-3 Vision Language Models: Smaller, Faster, Stronger","code":"https://github.com/kyegomez/PALI3","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":0,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":16,"model":"DePlot+FlanPaLM (CoT)","metrics":{"1:1 Accuracy":"67.3"},"uses_additional_data":false,"paper_date":"2022-12-20","paper":"/paper/deplot-one-shot-visual-language-reasoning-by","paper_url":"https://arxiv.org/abs/2212.10505v2","paper_title":"DePlot: One-shot visual language reasoning by plot-to-table translation","code":"https://github.com/huggingface/transformers","n_code_links":1,"syntology":null},{"rank_in_archive_order":17,"model":"Qwen-VL-Chat","metrics":{"1:1 Accuracy":"66.3"},"uses_additional_data":true,"paper_date":"2023-08-24","paper":"/paper/qwen-vl-a-frontier-large-vision-language","paper_url":"https://arxiv.org/abs/2308.12966v3","paper_title":"Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond","code":"https://github.com/qwenlm/qwen-vl","n_code_links":2,"syntology":{"n_ran":0,"n_unverified":2,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":18,"model":"UniChart","metrics":{"1:1 Accuracy":"66.24"},"uses_additional_data":true,"paper_date":"2023-05-24","paper":"/paper/unichart-a-universal-vision-language","paper_url":"https://arxiv.org/abs/2305.14761v3","paper_title":"UniChart: A Universal Vision-language Pretrained Model for Chart Comprehension and Reasoning","code":"https://github.com/vis-nlp/unichart","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":19,"model":"Qwen-VL","metrics":{"1:1 Accuracy":"65.7"},"uses_additional_data":true,"paper_date":"2023-08-24","paper":"/paper/qwen-vl-a-frontier-large-vision-language","paper_url":"https://arxiv.org/abs/2308.12966v3","paper_title":"Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond","code":"https://github.com/qwenlm/qwen-vl","n_code_links":2,"syntology":{"n_ran":0,"n_unverified":2,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":20,"model":"StructChart+GPT3.5 (STR ChartQA+SimChart9K)","metrics":{"1:1 Accuracy":"65.3"},"uses_additional_data":true,"paper_date":"2023-09-20","paper":"/paper/structchart-perception-structuring-reasoning","paper_url":"https://arxiv.org/abs/2309.11268v5","paper_title":"StructChart: On the Schema, Metric, and Augmentation for Visual Chart Understanding","code":"https://github.com/alpha-innovator/chartvlm","n_code_links":3,"syntology":null},{"rank_in_archive_order":21,"model":"MatCha","metrics":{"1:1 Accuracy":"64.2"},"uses_additional_data":false,"paper_date":"2022-12-19","paper":"/paper/matcha-enhancing-visual-language-pretraining","paper_url":"https://arxiv.org/abs/2212.09662v2","paper_title":"MatCha: Enhancing Visual Language Pretraining with Math Reasoning and Chart Derendering","code":"https://github.com/huggingface/transformers","n_code_links":1,"syntology":null},{"rank_in_archive_order":22,"model":"StructChart+GPT3.5 (STR)","metrics":{"1:1 Accuracy":"60.7"},"uses_additional_data":false,"paper_date":"2023-09-20","paper":"/paper/structchart-perception-structuring-reasoning","paper_url":"https://arxiv.org/abs/2309.11268v5","paper_title":"StructChart: On the Schema, Metric, and Augmentation for Visual Chart Understanding","code":"https://github.com/alpha-innovator/chartvlm","n_code_links":3,"syntology":null},{"rank_in_archive_order":23,"model":"Pix2Struct-large","metrics":{"1:1 Accuracy":"58.6"},"uses_additional_data":false,"paper_date":"2022-10-07","paper":"/paper/pix2struct-screenshot-parsing-as-pretraining","paper_url":"https://arxiv.org/abs/2210.03347v2","paper_title":"Pix2Struct: Screenshot Parsing as Pretraining for Visual Language Understanding","code":"https://github.com/huggingface/transformers","n_code_links":4,"syntology":{"n_ran":0,"n_unverified":5,"n_samples":5,"n_pointer_only_licence":0}},{"rank_in_archive_order":24,"model":"Pix2Struct-base","metrics":{"1:1 Accuracy":"56.0"},"uses_additional_data":false,"paper_date":"2022-10-07","paper":"/paper/pix2struct-screenshot-parsing-as-pretraining","paper_url":"https://arxiv.org/abs/2210.03347v2","paper_title":"Pix2Struct: Screenshot Parsing as Pretraining for Visual Language Understanding","code":"https://github.com/huggingface/transformers","n_code_links":4,"syntology":{"n_ran":0,"n_unverified":5,"n_samples":5,"n_pointer_only_licence":0}},{"rank_in_archive_order":25,"model":"VisionTapas-OCR","metrics":{"1:1 Accuracy":"45.5"},"uses_additional_data":false,"paper_date":"2022-03-19","paper":"/paper/chartqa-a-benchmark-for-question-answering","paper_url":"https://arxiv.org/abs/2203.10244v1","paper_title":"ChartQA: A Benchmark for Question Answering about Charts with Visual and Logical Reasoning","code":"https://github.com/vis-nlp/chartqa","n_code_links":1,"syntology":null},{"rank_in_archive_order":26,"model":"DePlot+GPT3 (Self-Consistency)","metrics":{"1:1 Accuracy":"42.3"},"uses_additional_data":false,"paper_date":"2022-12-20","paper":"/paper/deplot-one-shot-visual-language-reasoning-by","paper_url":"https://arxiv.org/abs/2212.10505v2","paper_title":"DePlot: One-shot visual language reasoning by plot-to-table translation","code":"https://github.com/huggingface/transformers","n_code_links":1,"syntology":null},{"rank_in_archive_order":27,"model":"DePlot+GPT3 (CoT)","metrics":{"1:1 Accuracy":"36.9"},"uses_additional_data":false,"paper_date":"2022-12-20","paper":"/paper/deplot-one-shot-visual-language-reasoning-by","paper_url":"https://arxiv.org/abs/2212.10505v2","paper_title":"DePlot: One-shot visual language reasoning by plot-to-table translation","code":"https://github.com/huggingface/transformers","n_code_links":1,"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 6,264 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":6264,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":3316},"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":10,"rows_with_any_sample_ran":6,"distinct_papers_with_graph_line":5,"distinct_papers_with_any_sample_ran":3,"samples_over_distinct_papers":{"n_ran":10,"n_unverified":8,"n_samples":18,"n_pointer_only_licence":2,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":25,"n_unverified":17,"n_samples":42,"n_pointer_only_licence":4,"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"}}}