{"url":"/sota/video-super-resolution-on-msu-vsr-benchmark","task":{"name":"Video Super-Resolution","url":"/task/video-super-resolution","note":null},"dataset":{"name":"MSU Video Super Resolution Benchmark: Detail Restoration","url":"/dataset/msu-vsr-benchmark"},"category":"Computer Vision","categories":["Computer Vision"],"category_note":null,"description":"**Video Super-Resolution** is a computer vision task that aims to increase the resolution of a video sequence, typically from lower to higher resolutions. The goal is to generate high-resolution video frames from low-resolution input, improving the overall quality of the video.\r\n\r\n<span style=\"color:grey; opacity: 0.6\">( Image credit: [Detail-revealing Deep Video Super-Resolution](https://github.com/jiangsutx/SPMC_VideoSR) )</span>","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":["Subjective score","ERQAv1.0","1 - LPIPS","SSIM","QRCRv1.0","PSNR","FPS"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"Subjective score":"higher","ERQAv1.0":null,"1 - LPIPS":null,"SSIM":"higher","QRCRv1.0":null,"PSNR":"higher","FPS":null}},"counts":{"rows":32,"rows_with_code":30,"rows_with_paper_page":30,"rows_dated":30,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"VRT","metrics":{"1 - LPIPS":"0.929","ERQAv1.0":"0.758","FPS":"2.778","PSNR":"31.669","QRCRv1.0":"0.722","SSIM":"0.902","Subjective score":"7.628"},"uses_additional_data":false,"paper_date":"2022-01-28","paper":"/paper/vrt-a-video-restoration-transformer","paper_url":"https://arxiv.org/abs/2201.12288v2","paper_title":"VRT: A Video Restoration Transformer","code":"https://github.com/jingyunliang/vrt","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":1,"n_samples":5,"n_pointer_only_licence":5}},{"rank_in_archive_order":2,"model":"BasicVSR","metrics":{"1 - LPIPS":"0.934","ERQAv1.0":"0.75","FPS":"2.128","PSNR":"31.443","QRCRv1.0":"0.709","SSIM":"0.9","Subjective score":"7.186"},"uses_additional_data":false,"paper_date":"2020-12-03","paper":"/paper/basicvsr-the-search-for-essential-components","paper_url":"https://arxiv.org/abs/2012.02181v2","paper_title":"BasicVSR: The Search for Essential Components in Video Super-Resolution and Beyond","code":"https://github.com/XPixelGroup/BasicSR","n_code_links":6,"syntology":null},{"rank_in_archive_order":3,"model":"RBPN","metrics":{"1 - LPIPS":"0.74","ERQAv1.0":"0.746","FPS":"0.043","PSNR":"31.407","QRCRv1.0":"0.629","SSIM":"0.899","Subjective score":"7.068"},"uses_additional_data":false,"paper_date":"2019-03-25","paper":"/paper/recurrent-back-projection-network-for-video","paper_url":"http://arxiv.org/abs/1903.10128v1","paper_title":"Recurrent Back-Projection Network for Video Super-Resolution","code":"https://github.com/alterzero/RBPN-PyTorch","n_code_links":7,"syntology":{"n_ran":6,"n_unverified":1,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":4,"model":"DBVSR","metrics":{"1 - LPIPS":"0.921","ERQAv1.0":"0.737","FPS":"0.241","PSNR":"31.071","QRCRv1.0":"0.629","SSIM":"0.894","Subjective score":"6.947"},"uses_additional_data":false,"paper_date":"2020-03-10","paper":"/paper/deep-blind-video-super-resolution","paper_url":"https://arxiv.org/abs/2003.04716v1","paper_title":"Deep Blind Video Super-resolution","code":"https://github.com/csbhr/Deep-Blind-VSR","n_code_links":2,"syntology":{"n_ran":10,"n_unverified":5,"n_samples":15,"n_pointer_only_licence":0}},{"rank_in_archive_order":5,"model":"iSeeBetter","metrics":{"1 - LPIPS":"0.741","ERQAv1.0":"0.748","FPS":"0.045","PSNR":"31.104","QRCRv1.0":"0.629","SSIM":"0.896","Subjective score":"6.809"},"uses_additional_data":false,"paper_date":"2020-06-21","paper":"/paper/iseebetter-spatio-temporal-video-super","paper_url":"https://arxiv.org/abs/2006.11161","paper_title":"iSeeBetter: Spatio-Temporal Video Super Resolution using Recurrent-Generative Back-Projection Networks","code":"https://github.com/amanchadha/iSeeBetter","n_code_links":1,"syntology":null},{"rank_in_archive_order":6,"model":"LGFN","metrics":{"1 - LPIPS":"0.903","ERQAv1.0":"0.74","FPS":"0.667","PSNR":"31.291","QRCRv1.0":"0.629","SSIM":"0.898","Subjective score":"6.505"},"uses_additional_data":false,"paper_date":"2020-09-22","paper":"/paper/local-global-fusion-network-for-video-super","paper_url":"https://ieeexplore.ieee.org/document/9203860/authors#authors","paper_title":"Local-Global Fusion Network for Video Super-Resolution","code":"https://github.com/BIOINSu/LGFN","n_code_links":1,"syntology":null},{"rank_in_archive_order":7,"model":"DynaVSR-R","metrics":{"1 - LPIPS":"0.884","ERQAv1.0":"0.709","FPS":"0.177","PSNR":"28.377","QRCRv1.0":"0.557","SSIM":"0.865","Subjective score":"6.136"},"uses_additional_data":false,"paper_date":"2020-11-09","paper":"/paper/dynavsr-dynamic-adaptive-blind-video-super","paper_url":"https://arxiv.org/abs/2011.04482v1","paper_title":"DynaVSR: Dynamic Adaptive Blind Video Super-Resolution","code":"https://github.com/esw0116/DynaVSR","n_code_links":1,"syntology":null},{"rank_in_archive_order":8,"model":"TMNet","metrics":{"1 - LPIPS":"0.931","ERQAv1.0":"0.712","FPS":"1.136","PSNR":"30.364","QRCRv1.0":"0.549","SSIM":"0.885","Subjective score":"6"},"uses_additional_data":false,"paper_date":"2021-04-21","paper":"/paper/temporal-modulation-network-for-controllable","paper_url":"https://arxiv.org/abs/2104.10642v2","paper_title":"Temporal Modulation Network for Controllable Space-Time Video Super-Resolution","code":"https://github.com/CS-GangXu/TMNet","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":5,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":9,"model":"COMISR","metrics":{"1 - LPIPS":"0.879","ERQAv1.0":"0.654","FPS":"1.613","PSNR":"26.708","QRCRv1.0":"0.619","SSIM":"0.84","Subjective score":"5.637"},"uses_additional_data":false,"paper_date":"2021-05-04","paper":"/paper/comisr-compression-informed-video-super","paper_url":"https://arxiv.org/abs/2105.01237v2","paper_title":"COMISR: Compression-Informed Video Super-Resolution","code":"https://github.com/google-research/google-research","n_code_links":2,"syntology":{"n_ran":0,"n_unverified":3,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":10,"model":"RSDN","metrics":{"1 - LPIPS":"0.819","ERQAv1.0":"0.667","FPS":"1.961","PSNR":"25.321","QRCRv1.0":"0.619","SSIM":"0.826","Subjective score":"5.566"},"uses_additional_data":false,"paper_date":"2020-08-02","paper":"/paper/video-super-resolution-with-recurrent","paper_url":"https://arxiv.org/abs/2008.00455v1","paper_title":"Video Super-Resolution with Recurrent Structure-Detail Network","code":"https://github.com/junpan19/RSDN","n_code_links":2,"syntology":null},{"rank_in_archive_order":11,"model":"TGA","metrics":{"1 - LPIPS":"0.859","ERQAv1.0":"0.669","FPS":"0.706","PSNR":"25.786","QRCRv1.0":"0.549","SSIM":"0.831","Subjective score":"5.529"},"uses_additional_data":false,"paper_date":"2020-07-21","paper":"/paper/video-super-resolution-with-temporal-group-1","paper_url":"https://arxiv.org/abs/2007.10595v1","paper_title":"Video Super-resolution with Temporal Group Attention","code":"https://github.com/junpan19/VSR_TGA","n_code_links":1,"syntology":null},{"rank_in_archive_order":12,"model":"TDAN","metrics":{"1 - LPIPS":"0.721","ERQAv1.0":"0.706","FPS":"0.493","PSNR":"30.244","QRCRv1.0":"0.609","SSIM":"0.883","Subjective score":"5.454"},"uses_additional_data":false,"paper_date":"2018-12-07","paper":"/paper/tdan-temporally-deformable-alignment-network","paper_url":"http://arxiv.org/abs/1812.02898v1","paper_title":"TDAN: Temporally Deformable Alignment Network for Video Super-Resolution","code":"https://github.com/YapengTian/TDAN_VSR","n_code_links":2,"syntology":null},{"rank_in_archive_order":13,"model":"Real-ESRGAN","metrics":{"1 - LPIPS":"0.895","ERQAv1.0":"0.663","FPS":"1.01","PSNR":"24.441","QRCRv1.0":"0","SSIM":"0.774","Subjective score":"5.392"},"uses_additional_data":false,"paper_date":"2021-07-22","paper":"/paper/real-esrgan-training-real-world-blind-super","paper_url":"https://arxiv.org/abs/2107.10833v2","paper_title":"Real-ESRGAN: Training Real-World Blind Super-Resolution with Pure Synthetic Data","code":"https://github.com/xinntao/Real-ESRGAN","n_code_links":8,"syntology":{"n_ran":6,"n_unverified":3,"n_samples":9,"n_pointer_only_licence":0}},{"rank_in_archive_order":14,"model":"ESRGAN","metrics":{"1 - LPIPS":"0.948","ERQAv1.0":"0.735","FPS":"1.004","PSNR":"27.33","QRCRv1.0":"0","SSIM":"0.808","Subjective score":"5.353"},"uses_additional_data":false,"paper_date":"2018-09-01","paper":"/paper/esrgan-enhanced-super-resolution-generative","paper_url":"http://arxiv.org/abs/1809.00219v2","paper_title":"ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks","code":"https://github.com/eriklindernoren/PyTorch-GAN","n_code_links":46,"syntology":{"n_ran":12,"n_unverified":32,"n_samples":44,"n_pointer_only_licence":0}},{"rank_in_archive_order":15,"model":"RRN-10L","metrics":{"1 - LPIPS":"0.842","ERQAv1.0":"0.627","FPS":"2.567","PSNR":"24.252","QRCRv1.0":"0.557","SSIM":"0.79","Subjective score":"5.35"},"uses_additional_data":false,"paper_date":"2020-08-13","paper":"/paper/revisiting-temporal-modeling-for-video-super","paper_url":"https://arxiv.org/abs/2008.05765v2","paper_title":"Revisiting Temporal Modeling for Video Super-resolution","code":"https://github.com/junpan19/RRN","n_code_links":2,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":16,"model":"DUF-28L","metrics":{"1 - LPIPS":"0.87","ERQAv1.0":"0.645","FPS":"0.418","PSNR":"25.852","QRCRv1.0":"0.549","SSIM":"0.83","Subjective score":"5.324"},"uses_additional_data":false,"paper_date":"2018-06-01","paper":"/paper/deep-video-super-resolution-network-using","paper_url":"http://openaccess.thecvf.com/content_cvpr_2018/html/Jo_Deep_Video_Super-Resolution_CVPR_2018_paper.html","paper_title":"Deep Video Super-Resolution Network Using Dynamic Upsampling Filters Without Explicit Motion Compensation","code":"https://github.com/yhjo09/VSR-DUF","n_code_links":1,"syntology":null},{"rank_in_archive_order":17,"model":"RealSR","metrics":{"1 - LPIPS":"0.911","ERQAv1.0":"0.69","FPS":"0.352","PSNR":"25.989","QRCRv1.0":"0","SSIM":"0.767","Subjective score":"5.286"},"uses_additional_data":false,"paper_date":"2020-06-19","paper":"/paper/real-world-super-resolution-via-kernel","paper_url":"https://ieeexplore.ieee.org/document/9150628","paper_title":"Real-World Super-Resolution via Kernel Estimation and Noise Injection","code":"https://github.com/nihui/realsr-ncnn-vulkan","n_code_links":2,"syntology":null},{"rank_in_archive_order":18,"model":"DUF-16L","metrics":{"1 - LPIPS":"0.868","ERQAv1.0":"0.641","FPS":"0.605","PSNR":"24.606","QRCRv1.0":"0.549","SSIM":"0.828","Subjective score":"5.124"},"uses_additional_data":false,"paper_date":"2018-06-01","paper":"/paper/deep-video-super-resolution-network-using","paper_url":"http://openaccess.thecvf.com/content_cvpr_2018/html/Jo_Deep_Video_Super-Resolution_CVPR_2018_paper.html","paper_title":"Deep Video Super-Resolution Network Using Dynamic Upsampling Filters Without Explicit Motion Compensation","code":"https://github.com/yhjo09/VSR-DUF","n_code_links":1,"syntology":null},{"rank_in_archive_order":19,"model":"D3Dnet","metrics":{"1 - LPIPS":"0.915","ERQAv1.0":"0.674","FPS":"0.041","PSNR":"29.703","QRCRv1.0":"0.549","SSIM":"0.876","Subjective score":"5.066"},"uses_additional_data":false,"paper_date":"2020-04-06","paper":"/paper/deformable-3d-convolution-for-video-super","paper_url":"https://arxiv.org/abs/2004.02803v5","paper_title":"Deformable 3D Convolution for Video Super-Resolution","code":"https://github.com/XinyiYing/D3Dnet","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":0,"n_samples":3,"n_pointer_only_licence":1}},{"rank_in_archive_order":20,"model":"RRN-5L","metrics":{"1 - LPIPS":"0.856","ERQAv1.0":"0.617","FPS":"2.74","PSNR":"23.786","QRCRv1.0":"0.549","SSIM":"0.789","Subjective score":"5.02"},"uses_additional_data":false,"paper_date":"2020-08-13","paper":"/paper/revisiting-temporal-modeling-for-video-super","paper_url":"https://arxiv.org/abs/2008.05765v2","paper_title":"Revisiting Temporal Modeling for Video Super-resolution","code":"https://github.com/junpan19/RRN","n_code_links":2,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":21,"model":"SOF-VSR-BD","metrics":{"1 - LPIPS":"0.895","ERQAv1.0":"0.647","FPS":"0.699","PSNR":"25.986","QRCRv1.0":"0.557","SSIM":"0.831","Subjective score":"4.863"},"uses_additional_data":false,"paper_date":"2020-01-06","paper":"/paper/deep-video-super-resolution-using-hr-optical","paper_url":"https://arxiv.org/abs/2001.02129v1","paper_title":"Deep Video Super-Resolution using HR Optical Flow Estimation","code":"https://github.com/LongguangWang/SOF-VSR","n_code_links":2,"syntology":null},{"rank_in_archive_order":22,"model":"SOF-VSR-BI","metrics":{"1 - LPIPS":"0.904","ERQAv1.0":"0.66","FPS":"0.571","PSNR":"29.381","QRCRv1.0":"0.557","SSIM":"0.872","Subjective score":"4.805"},"uses_additional_data":false,"paper_date":"2020-01-06","paper":"/paper/deep-video-super-resolution-using-hr-optical","paper_url":"https://arxiv.org/abs/2001.02129v1","paper_title":"Deep Video Super-Resolution using HR Optical Flow Estimation","code":"https://github.com/LongguangWang/SOF-VSR","n_code_links":2,"syntology":null},{"rank_in_archive_order":23,"model":"SwinIR","metrics":{"1 - LPIPS":"0.895","ERQAv1.0":"0.618","FPS":"0.407","PSNR":"25.12","QRCRv1.0":"0","SSIM":"0.782","Subjective score":"4.799"},"uses_additional_data":false,"paper_date":"2021-08-23","paper":"/paper/swinir-image-restoration-using-swin","paper_url":"https://arxiv.org/abs/2108.10257v1","paper_title":"SwinIR: Image Restoration Using Swin Transformer","code":"https://github.com/XPixelGroup/BasicSR","n_code_links":9,"syntology":{"n_ran":30,"n_unverified":15,"n_samples":45,"n_pointer_only_licence":5}},{"rank_in_archive_order":24,"model":"DynaVSR-V","metrics":{"1 - LPIPS":"0.859","ERQAv1.0":"0.643","FPS":"0.15","PSNR":"29.011","QRCRv1.0":"0.549","SSIM":"0.864","Subjective score":"4.359"},"uses_additional_data":false,"paper_date":"2020-11-09","paper":"/paper/dynavsr-dynamic-adaptive-blind-video-super","paper_url":"https://arxiv.org/abs/2011.04482v1","paper_title":"DynaVSR: Dynamic Adaptive Blind Video Super-Resolution","code":"https://github.com/esw0116/DynaVSR","n_code_links":1,"syntology":null},{"rank_in_archive_order":25,"model":"HCFlow","metrics":{"1 - LPIPS":"0.923","ERQAv1.0":"0.713","FPS":"0.066","PSNR":"26.067","QRCRv1.0":"0","SSIM":"0.791","Subjective score":"4.262"},"uses_additional_data":false,"paper_date":"2021-08-11","paper":"/paper/hierarchical-conditional-flow-a-unified","paper_url":"https://arxiv.org/abs/2108.05301v1","paper_title":"Hierarchical Conditional Flow: A Unified Framework for Image Super-Resolution and Image Rescaling","code":"https://github.com/jingyunliang/hcflow","n_code_links":1,"syntology":{"n_ran":5,"n_unverified":2,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":26,"model":"Real-ESRnet","metrics":{"1 - LPIPS":"0.871","ERQAv1.0":"0.598","FPS":"1.019","PSNR":"27.195","QRCRv1.0":"0","SSIM":"0.824","Subjective score":"3.697"},"uses_additional_data":false,"paper_date":"2021-07-22","paper":"/paper/real-esrgan-training-real-world-blind-super","paper_url":"https://arxiv.org/abs/2107.10833v2","paper_title":"Real-ESRGAN: Training Real-World Blind Super-Resolution with Pure Synthetic Data","code":"https://github.com/xinntao/Real-ESRGAN","n_code_links":8,"syntology":{"n_ran":6,"n_unverified":3,"n_samples":9,"n_pointer_only_licence":0}},{"rank_in_archive_order":27,"model":"SRMD","metrics":{"1 - LPIPS":"0.877","ERQAv1.0":"0.594","FPS":"5.882","PSNR":"27.672","QRCRv1.0":"0","SSIM":"0.834","Subjective score":"3.468"},"uses_additional_data":false,"paper_date":"2017-12-17","paper":"/paper/learning-a-single-convolutional-super","paper_url":"http://arxiv.org/abs/1712.06116v2","paper_title":"Learning a Single Convolutional Super-Resolution Network for Multiple Degradations","code":"https://github.com/cszn/SRMD","n_code_links":1,"syntology":null},{"rank_in_archive_order":28,"model":"waifu2x-cunet","metrics":{"1 - LPIPS":"0.861","ERQAv1.0":"0.6","FPS":"1.282","PSNR":"27.716","QRCRv1.0":"0","SSIM":"0.838","Subjective score":"3.308"},"uses_additional_data":false,"paper_date":null,"paper":null,"paper_url":null,"paper_title":"","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":29,"model":"waifu2x-anime","metrics":{"1 - LPIPS":"0.856","ERQAv1.0":"0.57","FPS":"3.125","PSNR":"27.176","QRCRv1.0":"0","SSIM":"0.819","Subjective score":"2.739"},"uses_additional_data":false,"paper_date":null,"paper":null,"paper_url":null,"paper_title":"","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":30,"model":"GFPGAN","metrics":{"1 - LPIPS":"0.793","ERQAv1.0":"0.538","FPS":"1.562","PSNR":"24.195","QRCRv1.0":"0","SSIM":"0.745","Subjective score":"2.686"},"uses_additional_data":false,"paper_date":"2021-01-11","paper":"/paper/towards-real-world-blind-face-restoration","paper_url":"https://arxiv.org/abs/2101.04061v2","paper_title":"Towards Real-World Blind Face Restoration with Generative Facial Prior","code":"https://github.com/TencentARC/GFPGAN","n_code_links":1,"syntology":null},{"rank_in_archive_order":31,"model":"ESPCN","metrics":{"1 - LPIPS":"0.765","ERQAv1.0":"0.521","FPS":"3.333","PSNR":"26.714","QRCRv1.0":"0","SSIM":"0.811","Subjective score":"2.099"},"uses_additional_data":false,"paper_date":"2016-09-16","paper":"/paper/real-time-single-image-and-video-super","paper_url":"http://arxiv.org/abs/1609.05158v2","paper_title":"Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network","code":"https://github.com/alexjc/neural-enhance","n_code_links":47,"syntology":{"n_ran":4,"n_unverified":16,"n_samples":20,"n_pointer_only_licence":1}},{"rank_in_archive_order":32,"model":"DFDnet","metrics":{"1 - LPIPS":"0.623","ERQAv1.0":"0.339","FPS":"0.909","PSNR":"24.832","QRCRv1.0":"0","SSIM":"0.759","Subjective score":"0.277"},"uses_additional_data":false,"paper_date":"2020-08-02","paper":"/paper/blind-face-restoration-via-deep-multi-scale","paper_url":"https://arxiv.org/abs/2008.00418v1","paper_title":"Blind Face Restoration via Deep Multi-scale Component Dictionaries","code":"https://github.com/csxmli2016/DFDNet","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":1,"n_samples":2,"n_pointer_only_licence":2}}],"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 6,885 of the 9,623 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":9623,"papers_checked":6885,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":2737},"order":"newest first by month (arXiv date, else the arXiv-id month), then arXiv id descending","columns":[],"entries":[]},"syntology":{"read_at":"2026-09-25T09:33:49+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":15,"rows_with_any_sample_ran":14,"distinct_papers_with_graph_line":13,"distinct_papers_with_any_sample_ran":12,"samples_over_distinct_papers":{"n_ran":84,"n_unverified":84,"n_samples":168,"n_pointer_only_licence":14,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":92,"n_unverified":87,"n_samples":179,"n_pointer_only_licence":14,"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"}}}