{"url":"/sota/motion-synthesis-on-humanml3d","task":{"name":"Motion Synthesis","url":"/task/motion-synthesis","note":null},"dataset":{"name":"HumanML3D","url":"/dataset/humanml3d"},"category":"Computer Vision","categories":["Computer Code","Computer Vision"],"category_note":null,"description":"Creating a video where people in the images move (such as blinking or smiling) requires specialized AI technology like Deepfake or Motion Synthesis, which I cannot do directly here.  \r\n\r\nHowever, if you want a video made from the images you provided, I can create an animated slideshow with cute effects and royalty-free background music. Would you be interested in this?","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":["FID","R Precision Top3","Diversity","Multimodality"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"FID":"lower","R Precision Top3":"higher","Diversity":null,"Multimodality":null}},"counts":{"rows":37,"rows_with_code":29,"rows_with_paper_page":37,"rows_dated":37,"rows_using_additional_data":2},"rows":[{"rank_in_archive_order":1,"model":"Motion Anything","metrics":{"Diversity":"9.521","FID":"0.028","Multimodality":"2.705","R Precision Top3":"0.829"},"uses_additional_data":false,"paper_date":"2025-03-10","paper":"/paper/motion-anything-any-to-motion-generation","paper_url":"https://arxiv.org/abs/2503.06955v1","paper_title":"Motion Anything: Any to Motion Generation","code":"https://github.com/steve-zeyu-zhang/MotionAnything","n_code_links":1,"syntology":null},{"rank_in_archive_order":2,"model":"MotionPCM","metrics":{"Diversity":" 9.575","FID":"0.030","Multimodality":" 1.714","R Precision Top3":"0.842"},"uses_additional_data":false,"paper_date":"2025-01-31","paper":"/paper/motionpcm-real-time-motion-synthesis-with","paper_url":"https://arxiv.org/abs/2501.19083v1","paper_title":"MotionPCM: Real-Time Motion Synthesis with Phased Consistency Model","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":3,"model":"BiPO","metrics":{"Diversity":"9.556","FID":"0.030","Multimodality":"1.374","R Precision Top3":"0.809"},"uses_additional_data":false,"paper_date":"2024-11-28","paper":"/paper/bipo-bidirectional-partial-occlusion-network-1","paper_url":"https://arxiv.org/abs/2412.00112v2","paper_title":"BiPO: Bidirectional Partial Occlusion Network for Text-to-Motion Synthesis","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":4,"model":"DisCoRD (+MoMask)","metrics":{"FID":"0.032","Multimodality":"1.288","R Precision Top3":"0.809"},"uses_additional_data":false,"paper_date":"2024-11-29","paper":"/paper/discord-discrete-tokens-to-continuous-motion","paper_url":"https://arxiv.org/abs/2411.19527v3","paper_title":"DisCoRD: Discrete Tokens to Continuous Motion via Rectified Flow Decoding","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":5,"model":"MoMask","metrics":{"FID":"0.045","Multimodality":"1.241","R Precision Top3":"0.807"},"uses_additional_data":false,"paper_date":"2023-11-29","paper":"/paper/momask-generative-masked-modeling-of-3d-human","paper_url":"https://arxiv.org/abs/2312.00063v1","paper_title":"MoMask: Generative Masked Modeling of 3D Human Motions","code":"https://github.com/EricGuo5513/momask-codes","n_code_links":1,"syntology":{"n_ran":8,"n_unverified":3,"n_samples":11,"n_pointer_only_licence":0}},{"rank_in_archive_order":6,"model":"BAD (CBS)","metrics":{"Diversity":"9.688","FID":"0.049","Multimodality":"1.119","R Precision Top3":"0.800"},"uses_additional_data":false,"paper_date":"2024-09-17","paper":"/paper/bad-bidirectional-auto-regressive-diffusion","paper_url":"https://arxiv.org/abs/2409.10847v1","paper_title":"BAD: Bidirectional Auto-regressive Diffusion for Text-to-Motion Generation","code":"https://github.com/rohollahhs/bad","n_code_links":1,"syntology":null},{"rank_in_archive_order":7,"model":"Free-T2M (StableMoFusion)","metrics":{"Diversity":"9.480","FID":"0.051","R Precision Top3":"0.803"},"uses_additional_data":false,"paper_date":"2025-01-30","paper":"/paper/free-t2m-frequency-enhanced-text-to-motion","paper_url":"https://arxiv.org/abs/2501.18232v1","paper_title":"Free-T2M: Frequency Enhanced Text-to-Motion Diffusion Model With Consistency Loss","code":"https://github.com/Hxxxz0/Free-T2m","n_code_links":1,"syntology":null},{"rank_in_archive_order":8,"model":"MCM","metrics":{"Diversity":"9.585","FID":"0.053","Multimodality":"0.8104","R Precision Top3":"0.788"},"uses_additional_data":false,"paper_date":"2024-04-19","paper":"/paper/mcm-multi-condition-motion-synthesis-1","paper_url":"https://arxiv.org/abs/2404.12886v1","paper_title":"MCM: Multi-condition Motion Synthesis Framework","code":"https://github.com/fluide1022/MCM","n_code_links":1,"syntology":null},{"rank_in_archive_order":9,"model":"BAMM","metrics":{"Diversity":"9.717","FID":"0.055","Multimodality":"1.687","R Precision Top3":"0.814"},"uses_additional_data":false,"paper_date":"2024-03-28","paper":"/paper/bamm-bidirectional-autoregressive-motion","paper_url":"https://arxiv.org/abs/2403.19435v3","paper_title":"BAMM: Bidirectional Autoregressive Motion Model","code":"https://github.com/exitudio/BAMM","n_code_links":1,"syntology":null},{"rank_in_archive_order":10,"model":"BAD (OAAS)","metrics":{"Diversity":"9.694","FID":"0.065","Multimodality":"1.194","R Precision Top3":"0.808"},"uses_additional_data":false,"paper_date":"2024-09-17","paper":"/paper/bad-bidirectional-auto-regressive-diffusion","paper_url":"https://arxiv.org/abs/2409.10847v1","paper_title":"BAD: Bidirectional Auto-regressive Diffusion for Text-to-Motion Generation","code":"https://github.com/rohollahhs/bad","n_code_links":1,"syntology":null},{"rank_in_archive_order":11,"model":"DiverseMotion (s=1)","metrics":{"Diversity":"9.551","FID":"0.070","Multimodality":"2.062","R Precision Top3":"0.783"},"uses_additional_data":true,"paper_date":"2023-09-04","paper":"/paper/diversemotion-towards-diverse-human-motion","paper_url":"https://arxiv.org/abs/2309.01372v1","paper_title":"DiverseMotion: Towards Diverse Human Motion Generation via Discrete Diffusion","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":12,"model":"DiverseMotion (s=2)","metrics":{"Diversity":"9.683","FID":"0.072","Multimodality":"1.869","R Precision Top3":"0.802"},"uses_additional_data":true,"paper_date":"2023-09-04","paper":"/paper/diversemotion-towards-diverse-human-motion","paper_url":"https://arxiv.org/abs/2309.01372v1","paper_title":"DiverseMotion: Towards Diverse Human Motion Generation via Discrete Diffusion","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":13,"model":"MMM (predict length)","metrics":{"Diversity":"9.411","FID":"0.080","Multimodality":"1.164","R Precision Top3":"0.794"},"uses_additional_data":false,"paper_date":"2023-12-06","paper":"/paper/mmm-generative-masked-motion-model","paper_url":"https://arxiv.org/abs/2312.03596v2","paper_title":"MMM: Generative Masked Motion Model","code":"https://github.com/exitudio/MMM","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":14,"model":"MMM (gt length)","metrics":{"Diversity":"9.577","FID":"0.089","Multimodality":"1.226","R Precision Top3":"0.804"},"uses_additional_data":false,"paper_date":"2023-12-06","paper":"/paper/mmm-generative-masked-motion-model","paper_url":"https://arxiv.org/abs/2312.03596v2","paper_title":"MMM: Generative Masked Motion Model","code":"https://github.com/exitudio/MMM","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":15,"model":"ReMoDiffuse","metrics":{"Diversity":"9.018","FID":"0.103","Multimodality":"1.795","R Precision Top3":"0.795"},"uses_additional_data":false,"paper_date":"2023-04-03","paper":"/paper/remodiffuse-retrieval-augmented-motion","paper_url":"https://arxiv.org/abs/2304.01116v1","paper_title":"ReMoDiffuse: Retrieval-Augmented Motion Diffusion Model","code":"https://github.com/mingyuan-zhang/ReMoDiffuse","n_code_links":1,"syntology":null},{"rank_in_archive_order":16,"model":"ParCo","metrics":{"Diversity":"9.576","FID":"0.109","Multimodality":"1.382","R Precision Top3":"0.801"},"uses_additional_data":false,"paper_date":"2024-03-27","paper":"/paper/parco-part-coordinating-text-to-motion","paper_url":"https://arxiv.org/abs/2403.18512v2","paper_title":"ParCo: Part-Coordinating Text-to-Motion Synthesis","code":"https://github.com/qrzou/parco","n_code_links":1,"syntology":null},{"rank_in_archive_order":17,"model":"GUESS","metrics":{"Diversity":"9.826","FID":"0.109","Multimodality":"2.430","R Precision Top3":"0.787"},"uses_additional_data":false,"paper_date":"2024-01-04","paper":"/paper/guess-gradually-enriching-synthesis-for-text","paper_url":"https://arxiv.org/abs/2401.02142v2","paper_title":"GUESS:GradUally Enriching SyntheSis for Text-Driven Human Motion Generation","code":"https://github.com/xuehao-gao/guess","n_code_links":1,"syntology":null},{"rank_in_archive_order":18,"model":"AttT2M","metrics":{"Diversity":"9.700","FID":"0.112","Multimodality":"2.452","R Precision Top3":"0.786 "},"uses_additional_data":false,"paper_date":"2023-09-02","paper":"/paper/attt2m-text-driven-human-motion-generation-1","paper_url":"https://arxiv.org/abs/2309.00796v1","paper_title":"AttT2M: Text-Driven Human Motion Generation with Multi-Perspective Attention Mechanism","code":"https://github.com/zcymonkey/attt2m","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":"EMDM","metrics":{"Diversity":"9.551","FID":"0.112","Multimodality":"1.641","R Precision Top3":"0.786"},"uses_additional_data":false,"paper_date":"2023-12-04","paper":"/paper/emdm-efficient-motion-diffusion-model-for","paper_url":"https://arxiv.org/abs/2312.02256v3","paper_title":"EMDM: Efficient Motion Diffusion Model for Fast and High-Quality Motion Generation","code":"https://github.com/frank-zy-dou/emdm","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":4,"n_samples":7,"n_pointer_only_licence":7}},{"rank_in_archive_order":20,"model":"GraphMotion","metrics":{"Diversity":"9.692","FID":"0.116","Multimodality":"2.766","R Precision Top3":"0.785"},"uses_additional_data":false,"paper_date":"2023-09-21","paper":"/paper/act-as-you-wish-fine-grained-control-of","paper_url":"https://openreview.net/forum?id=xSEhb2j3TK","paper_title":"Act As You Wish: Fine-Grained Control of Motion Diffusion Model with Hierarchical Semantic Graphs","code":"https://github.com/jpthu17/graphmotion","n_code_links":1,"syntology":null},{"rank_in_archive_order":21,"model":"T2M-GPT (τ = 0.5)","metrics":{"Diversity":"9.761","FID":"0.116","Multimodality":"1.856","R Precision Top3":"0.775"},"uses_additional_data":false,"paper_date":"2023-01-15","paper":"/paper/t2m-gpt-generating-human-motion-from-textual","paper_url":"https://arxiv.org/abs/2301.06052v4","paper_title":"T2M-GPT: Generating Human Motion from Textual Descriptions with Discrete Representations","code":"https://github.com/Mael-zys/T2M-GPT","n_code_links":1,"syntology":{"n_ran":5,"n_unverified":2,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":22,"model":"T2M-GPT (τ = 0)","metrics":{"Diversity":"9.844","FID":"0.140","Multimodality":"3.285","R Precision Top3":"0.685"},"uses_additional_data":false,"paper_date":"2023-01-15","paper":"/paper/t2m-gpt-generating-human-motion-from-textual","paper_url":"https://arxiv.org/abs/2301.06052v4","paper_title":"T2M-GPT: Generating Human Motion from Textual Descriptions with Discrete Representations","code":"https://github.com/Mael-zys/T2M-GPT","n_code_links":1,"syntology":{"n_ran":5,"n_unverified":2,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":23,"model":"T2M-GPT (τ ∈ U[0, 1])","metrics":{"Diversity":"9.722","FID":"0.141","Multimodality":"1.831","R Precision Top3":"0.775"},"uses_additional_data":false,"paper_date":"2023-01-15","paper":"/paper/t2m-gpt-generating-human-motion-from-textual","paper_url":"https://arxiv.org/abs/2301.06052v4","paper_title":"T2M-GPT: Generating Human Motion from Textual Descriptions with Discrete Representations","code":"https://github.com/Mael-zys/T2M-GPT","n_code_links":1,"syntology":{"n_ran":5,"n_unverified":2,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":24,"model":"FineMoGen","metrics":{"Diversity":"9.263","FID":"0.151","Multimodality":"2.696","R Precision Top3":"0.784"},"uses_additional_data":false,"paper_date":"2023-12-22","paper":"/paper/finemogen-fine-grained-spatio-temporal-motion-1","paper_url":"https://arxiv.org/abs/2312.15004v1","paper_title":"FineMoGen: Fine-Grained Spatio-Temporal Motion Generation and Editing","code":"https://github.com/mingyuan-zhang/FineMoGen","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":0,"n_samples":3,"n_pointer_only_licence":3}},{"rank_in_archive_order":25,"model":"Fg-T2M","metrics":{"Diversity":"9.278","FID":"0.243","Multimodality":"1.614","R Precision Top3":"0.783"},"uses_additional_data":false,"paper_date":"2023-09-12","paper":"/paper/fg-t2m-fine-grained-text-driven-human-motion","paper_url":"https://arxiv.org/abs/2309.06284v1","paper_title":"Fg-T2M: Fine-Grained Text-Driven Human Motion Generation via Diffusion Model","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":26,"model":"Motion Mamba","metrics":{"Diversity":"9.871","FID":"0.281","Multimodality":"2.294","R Precision Top3":"0.792"},"uses_additional_data":false,"paper_date":"2024-03-12","paper":"/paper/motion-mamba-efficient-and-long-sequence","paper_url":"https://arxiv.org/abs/2403.07487v4","paper_title":"Motion Mamba: Efficient and Long Sequence Motion Generation","code":"https://github.com/steve-zeyu-zhang/MotionMamba","n_code_links":1,"syntology":null},{"rank_in_archive_order":27,"model":"MotionLCM (4-step)","metrics":{"Diversity":"9.607","FID":"0.304","Multimodality":"2.259","R Precision Top3":"0.798"},"uses_additional_data":false,"paper_date":"2024-04-30","paper":"/paper/motionlcm-real-time-controllable-motion","paper_url":"https://arxiv.org/abs/2404.19759v3","paper_title":"MotionLCM: Real-time Controllable Motion Generation via Latent Consistency Model","code":"https://github.com/Dai-Wenxun/MotionLCM","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":28,"model":"MLD","metrics":{"Diversity":"9.724","FID":"0.473","Multimodality":"2.413","R Precision Top3":"0.772"},"uses_additional_data":false,"paper_date":"2022-12-08","paper":"/paper/executing-your-commands-via-motion-diffusion","paper_url":"https://arxiv.org/abs/2212.04048v3","paper_title":"Executing your Commands via Motion Diffusion in Latent Space","code":"https://github.com/chenfengye/motion-latent-diffusion","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":29,"model":"MDM","metrics":{"Diversity":"9.559","FID":"0.544","Multimodality":"2.799","R Precision Top3":"0.611"},"uses_additional_data":false,"paper_date":"2022-09-29","paper":"/paper/human-motion-diffusion-model","paper_url":"https://arxiv.org/abs/2209.14916v2","paper_title":"Human Motion Diffusion Model","code":"https://github.com/guytevet/motion-diffusion-model","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":30,"model":"MotionDiffuse","metrics":{"Diversity":"9.410","FID":"0.630","Multimodality":"1.553","R Precision Top3":"0.782"},"uses_additional_data":false,"paper_date":"2022-08-31","paper":"/paper/motiondiffuse-text-driven-human-motion","paper_url":"https://arxiv.org/abs/2208.15001v1","paper_title":"MotionDiffuse: Text-Driven Human Motion Generation with Diffusion Model","code":"https://github.com/mingyuan-zhang/MotionDiffuse","n_code_links":2,"syntology":null},{"rank_in_archive_order":31,"model":"MAA","metrics":{"Diversity":"8.23","FID":"0.774","R Precision Top3":"0.676"},"uses_additional_data":false,"paper_date":"2023-05-16","paper":"/paper/make-an-animation-large-scale-text","paper_url":"https://arxiv.org/abs/2305.09662v1","paper_title":"Make-An-Animation: Large-Scale Text-conditional 3D Human Motion Generation","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":32,"model":"TM2D (t2m)","metrics":{"Diversity":"9.513","FID":"1.021","Multimodality":"4.139"},"uses_additional_data":false,"paper_date":"2023-04-05","paper":"/paper/tm2d-bimodality-driven-3d-dance-generation","paper_url":"https://arxiv.org/abs/2304.02419v2","paper_title":"TM2D: Bimodality Driven 3D Dance Generation via Music-Text Integration","code":"https://github.com/Garfield-kh/TM2D","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":33,"model":"T2M","metrics":{"Diversity":"9.175","FID":"1.087","Multimodality":"2.219","R Precision Top3":"0.736"},"uses_additional_data":false,"paper_date":"2022-01-01","paper":"/paper/generating-diverse-and-natural-3d-human","paper_url":"http://openaccess.thecvf.com//content/CVPR2022/html/Guo_Generating_Diverse_and_Natural_3D_Human_Motions_From_Text_CVPR_2022_paper.html","paper_title":"Generating Diverse and Natural 3D Human Motions From Text","code":"https://github.com/EricGuo5513/text-to-motion","n_code_links":1,"syntology":null},{"rank_in_archive_order":34,"model":"TM2T","metrics":{"Diversity":"8.589","FID":"1.501","Multimodality":"2.424","R Precision Top3":"0.729"},"uses_additional_data":false,"paper_date":"2022-07-04","paper":"/paper/tm2t-stochastic-and-tokenized-modeling-for","paper_url":"https://arxiv.org/abs/2207.01696v2","paper_title":"TM2T: Stochastic and Tokenized Modeling for the Reciprocal Generation of 3D Human Motions and Texts","code":"https://github.com/EricGuo5513/TM2T","n_code_links":1,"syntology":{"n_ran":5,"n_unverified":2,"n_samples":7,"n_pointer_only_licence":1}},{"rank_in_archive_order":35,"model":"Text2Gesture","metrics":{"Diversity":"6.409","FID":"5.012","R Precision Top3":"0.345"},"uses_additional_data":false,"paper_date":"2022-07-04","paper":"/paper/tm2t-stochastic-and-tokenized-modeling-for","paper_url":"https://arxiv.org/abs/2207.01696v2","paper_title":"TM2T: Stochastic and Tokenized Modeling for the Reciprocal Generation of 3D Human Motions and Texts","code":"https://github.com/EricGuo5513/TM2T","n_code_links":1,"syntology":{"n_ran":5,"n_unverified":2,"n_samples":7,"n_pointer_only_licence":1}},{"rank_in_archive_order":36,"model":"Diffuion Motion","metrics":{"Diversity":"23.692","FID":"10.21","R Precision Top3":"0.735"},"uses_additional_data":false,"paper_date":"2022-10-22","paper":"/paper/diffusion-motion-generate-text-guided-3d","paper_url":"https://arxiv.org/abs/2210.12315v2","paper_title":"Diffusion Motion: Generate Text-Guided 3D Human Motion by Diffusion Model","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":37,"model":"Language2Pose","metrics":{"Diversity":"7.676","FID":"11.02","R Precision Top3":"0.486"},"uses_additional_data":false,"paper_date":"2022-07-04","paper":"/paper/tm2t-stochastic-and-tokenized-modeling-for","paper_url":"https://arxiv.org/abs/2207.01696v2","paper_title":"TM2T: Stochastic and Tokenized Modeling for the Reciprocal Generation of 3D Human Motions and Texts","code":"https://github.com/EricGuo5513/TM2T","n_code_links":1,"syntology":{"n_ran":5,"n_unverified":2,"n_samples":7,"n_pointer_only_licence":1}}],"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":16,"rows_with_any_sample_ran":15,"distinct_papers_with_graph_line":11,"distinct_papers_with_any_sample_ran":10,"samples_over_distinct_papers":{"n_ran":29,"n_unverified":12,"n_samples":41,"n_pointer_only_licence":15,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":50,"n_unverified":20,"n_samples":70,"n_pointer_only_licence":18,"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"}}}