{"as_of":"2026-08-17T01:14:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:ba828314885758f813f5e86efd0cb6da3b9d854c104052141f2024d867b4ff4b","coverage":[{"denominator":21,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":21,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-14T13:59:54.983533Z","state":"measured"},{"denominator":22,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":22,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-16T06:30:59.297886+00:00","state":"measured"},{"denominator":1,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":1,"source":"paper_references, paper_reference_links","source_observed_at":"2026-05-13T09:41:37.937046Z","state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"arxiv_reference","source_observed_at":"2026-05-13T09:41:38.054219Z","state":"measured"}],"external_citation_measurements":[],"inbound":[{"citation":{"cited_paper":{"arxiv_id":"1908.08986","last_updated":"2019-08-12T08:27:49Z","snapshot_observed_at":"2026-08-14T16:19:54.488219Z","submitted_at":"2019-08-12T08:27:49Z","title":"Mix & Match: training convnets with mixed image sizes for improved accuracy, speed and scale resiliency","version":1},"cited_work":{"arxiv_id":"1908.08986","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"1908.08986","snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"preprint arXiv:1908.08986 , year=","venue":null,"work_id":"a10079b5-8b3a-430e-b24a-a98e7ef428e3","year":1908},"citing_paper":{"arxiv_id":"2309.16588","last_updated":"2024-04-12T09:38:33Z","snapshot_observed_at":"2026-08-07T09:40:32.614733Z","submitted_at":"2023-09-28T16:45:46Z","title":"Vision Transformers Need Registers","version":2},"reference_index":155,"source":"arxiv_source","source_observed_at":"2026-05-13T09:41:37.937046Z"},"links":{"cited_paper":"/paper/1908.08986","citing_paper":"/paper/2309.16588"},"observation_digest":"sha256:3a820b6bd8757d7f944aa5ee86f7b4f5ea3afff6c76baca7151bcb744d0877f0","observation_id":"eacbb2d3-22bb-4208-91fa-59ab85159a63","resolution":{"observed_at":"2026-05-13T09:41:38.056103Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}}],"links":{"evidence":"/evidence","html":"/paper/1908.08986/citation-record","integrity":"/paper/1908.08986/integrity","json":"/paper/1908.08986/citation-record.json","paper":"/paper/1908.08986"},"outbound":[{"citation":{"cited_paper":{"arxiv_id":"1805.09501","last_updated":"2019-04-11T22:39:27Z","snapshot_observed_at":"2026-08-14T19:11:44.108868Z","submitted_at":"2018-05-24T04:05:42Z","title":"AutoAugment: Learning Augmentation Policies from Data","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1805.09501","snapshot_observed_at":"2026-08-14T13:59:54.887694Z","title":"Autoaugment: Learning augmentation policies from data","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"1908.08986","last_updated":"2019-08-12T08:27:49Z","snapshot_observed_at":"2026-08-14T16:19:54.488219Z","submitted_at":"2019-08-12T08:27:49Z","title":"Mix & Match: training convnets with mixed image sizes for improved accuracy, speed and scale resiliency","version":1},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-14T13:59:54.887694Z"},"links":{"cited_paper":"/paper/1805.09501","citing_paper":"/paper/1908.08986"},"observation_digest":"sha256:d2784dda81ae7b30ec745a8d222bf16563a32ce2a3959544f46b2a2a4d49ca85","observation_id":"2ccadf29-9620-4f6c-8403-211f0b27b863","resolution":{"observed_at":"2026-08-14T13:59:54.887694Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1706.02677","last_updated":"2018-04-30T21:53:41Z","snapshot_observed_at":"2026-08-09T05:23:26.365677Z","submitted_at":"2017-06-08T16:51:53Z","title":"Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1706.02677","snapshot_observed_at":"2026-08-14T13:59:54.904189Z","title":"Accurate, large minibatch sgd: Training imagenet in 1 hour","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"1908.08986","last_updated":"2019-08-12T08:27:49Z","snapshot_observed_at":"2026-08-14T16:19:54.488219Z","submitted_at":"2019-08-12T08:27:49Z","title":"Mix & Match: training convnets with mixed image sizes for improved accuracy, speed and scale resiliency","version":1},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-14T13:59:54.904189Z"},"links":{"cited_paper":"/paper/1706.02677","citing_paper":"/paper/1908.08986"},"observation_digest":"sha256:67875bc1cc6333923009cf9ee906df766f1d667111d602ba319a758b33dcf247","observation_id":"db3ac107-5e16-44d7-bcd4-0f623981e8d4","resolution":{"observed_at":"2026-08-14T13:59:54.904189Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1901.09335","last_updated":"2019-01-27T08:02:26Z","snapshot_observed_at":"2026-08-14T17:25:24.489718Z","submitted_at":"2019-01-27T08:02:26Z","title":"Augment your batch: better training with larger batches","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1901.09335","snapshot_observed_at":"2026-08-14T13:59:54.909169Z","title":"Augment your batch: better training with larger batches","venue":null,"work_id":null,"year":1901},"citing_paper":{"arxiv_id":"1908.08986","last_updated":"2019-08-12T08:27:49Z","snapshot_observed_at":"2026-08-14T16:19:54.488219Z","submitted_at":"2019-08-12T08:27:49Z","title":"Mix & Match: training convnets with mixed image sizes for improved accuracy, speed and scale resiliency","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-14T13:59:54.909169Z"},"links":{"cited_paper":"/paper/1901.09335","citing_paper":"/paper/1908.08986"},"observation_digest":"sha256:2df031a87219f9656765209e51d7380bb12f48483b88658bbe72da9c1c98efbe","observation_id":"de8625e6-9143-4f58-b1e6-adda1e8bf258","resolution":{"observed_at":"2026-08-14T13:59:54.909169Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1312.5402","last_updated":"2013-12-19T04:23:23Z","snapshot_observed_at":"2026-08-14T23:51:04.407719Z","submitted_at":"2013-12-19T04:23:23Z","title":"Some Improvements on Deep Convolutional Neural Network Based Image Classification","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1312.5402","snapshot_observed_at":"2026-08-14T13:59:54.914028Z","title":"Some improvements on deep convolutional neural network based image classi- fication","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"1908.08986","last_updated":"2019-08-12T08:27:49Z","snapshot_observed_at":"2026-08-14T16:19:54.488219Z","submitted_at":"2019-08-12T08:27:49Z","title":"Mix & Match: training convnets with mixed image sizes for improved accuracy, speed and scale resiliency","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-14T13:59:54.914028Z"},"links":{"cited_paper":"/paper/1312.5402","citing_paper":"/paper/1908.08986"},"observation_digest":"sha256:724825b3843579ffd169a741af7557a0b9f9f84b2a1fe3b7550fb2994abb2f77","observation_id":"44209322-70f5-438b-994f-3d6c27b97cef","resolution":{"observed_at":"2026-08-14T13:59:54.914028Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-14T13:59:55.301592Z","title":"Batch renormalization: Towards reducing minibatch dependence in batch-normalized models","venue":null,"work_id":"becb66ff-4d98-44ca-9d63-488f9a07e405","year":1945},"citing_paper":{"arxiv_id":"1908.08986","last_updated":"2019-08-12T08:27:49Z","snapshot_observed_at":"2026-08-14T16:19:54.488219Z","submitted_at":"2019-08-12T08:27:49Z","title":"Mix & Match: training convnets with mixed image sizes for improved accuracy, speed and scale resiliency","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-14T13:59:54.928696Z"},"links":{"citing_paper":"/paper/1908.08986"},"observation_digest":"sha256:31f5511c375557e17a23ca1b02cbbb080202e6b0b0b7e4d484eb3080a86b60c1","observation_id":"e36ed28c-cbc9-41dd-85c8-4159e92943e4","resolution":{"observed_at":"2026-08-14T13:59:55.306873Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1608.03983","last_updated":"2017-05-03T16:28:09Z","snapshot_observed_at":"2026-07-06T05:06:55.589962Z","submitted_at":"2016-08-13T13:46:05Z","title":"SGDR: Stochastic Gradient Descent with Warm Restarts","version":5},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1608.03983","snapshot_observed_at":"2026-08-14T13:59:54.945101Z","title":"Sgdr: Stochastic gradient descent with warm restarts","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"1908.08986","last_updated":"2019-08-12T08:27:49Z","snapshot_observed_at":"2026-08-14T16:19:54.488219Z","submitted_at":"2019-08-12T08:27:49Z","title":"Mix & Match: training convnets with mixed image sizes for improved accuracy, speed and scale resiliency","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-14T13:59:54.945101Z"},"links":{"cited_paper":"/paper/1608.03983","citing_paper":"/paper/1908.08986"},"observation_digest":"sha256:7222a2786862f66a809a3209aabc210dfee9a08ffe2b1927601e22d9e2c944b3","observation_id":"3d308dda-4c7d-48e8-806b-2addad6bfbc3","resolution":{"observed_at":"2026-08-14T13:59:54.945101Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1811.03600","last_updated":"2019-07-19T00:47:44Z","snapshot_observed_at":"2026-08-14T18:02:01.603978Z","submitted_at":"2018-11-08T18:33:41Z","title":"Measuring the Effects of Data Parallelism on Neural Network Training","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1811.03600","snapshot_observed_at":"2026-08-14T13:59:54.949782Z","title":"Measuring the effects of data parallelism on neural network training","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"1908.08986","last_updated":"2019-08-12T08:27:49Z","snapshot_observed_at":"2026-08-14T16:19:54.488219Z","submitted_at":"2019-08-12T08:27:49Z","title":"Mix & Match: training convnets with mixed image sizes for improved accuracy, speed and scale resiliency","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-14T13:59:54.949782Z"},"links":{"cited_paper":"/paper/1811.03600","citing_paper":"/paper/1908.08986"},"observation_digest":"sha256:7d878e05a8afc390e9b762f32b4853f85da3e98a432bf1a0b451f3cec69d9f3c","observation_id":"7702db2e-bf59-4a06-a26a-db8b2aac3f9d","resolution":{"observed_at":"2026-08-14T13:59:54.949782Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1409.1556","last_updated":"2015-04-10T16:25:04Z","snapshot_observed_at":"2026-08-14T23:20:42.336514Z","submitted_at":"2014-09-04T19:48:04Z","title":"Very Deep Convolutional Networks for Large-Scale Image Recognition","version":6},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1409.1556","snapshot_observed_at":"2026-08-14T13:59:54.954562Z","title":"Very deep convolutional networks for large-scale image recognition","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"1908.08986","last_updated":"2019-08-12T08:27:49Z","snapshot_observed_at":"2026-08-14T16:19:54.488219Z","submitted_at":"2019-08-12T08:27:49Z","title":"Mix & Match: training convnets with mixed image sizes for improved accuracy, speed and scale resiliency","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-14T13:59:54.954562Z"},"links":{"cited_paper":"/paper/1409.1556","citing_paper":"/paper/1908.08986"},"observation_digest":"sha256:c7566be5e9bfdc92cd6c6e7f5543bd676ebe42885a5da8aa18d2c218d716a716","observation_id":"d97df3bd-70b5-474c-ace9-70d9c4c0b98b","resolution":{"observed_at":"2026-08-14T13:59:54.954562Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1906.06423","last_updated":"2022-01-20T11:02:01Z","snapshot_observed_at":"2026-08-14T16:15:19.784756Z","submitted_at":"2019-06-14T22:27:30Z","title":"Fixing the train-test resolution discrepancy","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1906.06423","snapshot_observed_at":"2026-08-14T13:59:54.963775Z","title":"A¨aron van den Oord, Sander Dieleman, Heiga Zen, Karen Simonyan, Oriol Vinyals, Alexander Graves, Nal Kalchbrenner, Andrew Senior, and Koray Kavukcuoglu","venue":null,"work_id":null,"year":1906},"citing_paper":{"arxiv_id":"1908.08986","last_updated":"2019-08-12T08:27:49Z","snapshot_observed_at":"2026-08-14T16:19:54.488219Z","submitted_at":"2019-08-12T08:27:49Z","title":"Mix & Match: training convnets with mixed image sizes for improved accuracy, speed and scale resiliency","version":1},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-14T13:59:54.963775Z"},"links":{"cited_paper":"/paper/1906.06423","citing_paper":"/paper/1908.08986"},"observation_digest":"sha256:da1de31255e01bba5820a0ea4d7c57de53bf3c718d74164e84367b5e48f37ba1","observation_id":"e32ec44a-21b0-488e-9225-087865a55379","resolution":{"observed_at":"2026-08-14T13:59:54.963775Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1708.03888","last_updated":"2017-09-13T23:25:07Z","snapshot_observed_at":"2026-08-14T20:41:41.185318Z","submitted_at":"2017-08-13T11:01:57Z","title":"Large Batch Training of Convolutional Networks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1708.03888","snapshot_observed_at":"2026-08-14T13:59:54.973605Z","title":"Large batch training of convolutional networks.arXiv preprint arXiv:1708.03888,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"1908.08986","last_updated":"2019-08-12T08:27:49Z","snapshot_observed_at":"2026-08-14T16:19:54.488219Z","submitted_at":"2019-08-12T08:27:49Z","title":"Mix & Match: training convnets with mixed image sizes for improved accuracy, speed and scale resiliency","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-14T13:59:54.973605Z"},"links":{"cited_paper":"/paper/1708.03888","citing_paper":"/paper/1908.08986"},"observation_digest":"sha256:8d756b7afe2b37b99e902aff85ed31c34a94b154474d99e1e5ddf05b559fa109","observation_id":"967b7a0d-5e77-4184-b921-804428cd34e0","resolution":{"observed_at":"2026-08-14T13:59:54.973605Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-14T13:59:55.286185Z","title":null,"venue":null,"work_id":"fe9b21d0-58b6-4e51-bd25-91ffcf3de69a","year":2016},"citing_paper":{"arxiv_id":"1908.08986","last_updated":"2019-08-12T08:27:49Z","snapshot_observed_at":"2026-08-14T16:19:54.488219Z","submitted_at":"2019-08-12T08:27:49Z","title":"Mix & Match: training convnets with mixed image sizes for improved accuracy, speed and scale resiliency","version":1},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-14T13:59:54.978952Z"},"links":{"citing_paper":"/paper/1908.08986"},"observation_digest":"sha256:b63434a86dceeb48ba2c604a5456bf684759fd089736d78062ce82bdf4f3a3f7","observation_id":"028ae5b1-52ca-408a-bf89-6f62586a30fe","resolution":{"observed_at":"2026-08-14T13:59:55.291032Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-14T13:59:55.270840Z","title":"Solid lines are test errors while dotted lines are for training","venue":null,"work_id":"d494a5f5-3b4e-4a00-8b5f-3f22554ab975","year":2018},"citing_paper":{"arxiv_id":"1908.08986","last_updated":"2019-08-12T08:27:49Z","snapshot_observed_at":"2026-08-14T16:19:54.488219Z","submitted_at":"2019-08-12T08:27:49Z","title":"Mix & Match: training convnets with mixed image sizes for improved accuracy, speed and scale resiliency","version":1},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-14T13:59:54.983533Z"},"links":{"citing_paper":"/paper/1908.08986"},"observation_digest":"sha256:05099428de3333da47fb851ffaab171bdfee9cb5ea30da9c3585e0b3e88d2385","observation_id":"54aa3dd1-75fc-4220-81a2-2dd957128702","resolution":{"observed_at":"2026-08-14T13:59:55.275600Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1708.04552","last_updated":"2017-11-29T14:51:40Z","snapshot_observed_at":"2026-08-13T11:54:46.030796Z","submitted_at":"2017-08-15T15:21:53Z","title":"Improved Regularization of Convolutional Neural Networks with Cutout","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1708.04552","snapshot_observed_at":"2026-08-14T13:59:54.892804Z","title":"Improved regularization of convolutional neural networks with cutout","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"1908.08986","last_updated":"2019-08-12T08:27:49Z","snapshot_observed_at":"2026-08-14T16:19:54.488219Z","submitted_at":"2019-08-12T08:27:49Z","title":"Mix & Match: training convnets with mixed image sizes for improved accuracy, speed and scale resiliency","version":1},"reference_index":2009,"source":"pdf_text","source_observed_at":"2026-08-14T13:59:54.892804Z"},"links":{"cited_paper":"/paper/1708.04552","citing_paper":"/paper/1908.08986"},"observation_digest":"sha256:b5c68b474a191640fa1d89c51481c2591803e3c7f186dc2c5c4dad574c63136a","observation_id":"0ebd4503-808e-4642-bece-04a29071ec5f","resolution":{"observed_at":"2026-08-14T13:59:54.892804Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1312.4400","last_updated":"2014-03-04T05:15:42Z","snapshot_observed_at":"2026-08-14T23:51:30.531758Z","submitted_at":"2013-12-16T15:34:13Z","title":"Network In Network","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1312.4400","snapshot_observed_at":"2026-08-14T13:59:54.940107Z","title":"Network in network","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"1908.08986","last_updated":"2019-08-12T08:27:49Z","snapshot_observed_at":"2026-08-14T16:19:54.488219Z","submitted_at":"2019-08-12T08:27:49Z","title":"Mix & Match: training convnets with mixed image sizes for improved accuracy, speed and scale resiliency","version":1},"reference_index":2012,"source":"pdf_text","source_observed_at":"2026-08-14T13:59:54.940107Z"},"links":{"cited_paper":"/paper/1312.4400","citing_paper":"/paper/1908.08986"},"observation_digest":"sha256:094db385f1b486e84591d280709d005b1a92584d4add6bbb17ef1ead4530150b","observation_id":"8732085e-d650-4ecf-a4ef-8c063a8f15a3","resolution":{"observed_at":"2026-08-14T13:59:54.940107Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-14T13:59:55.317442Z","title":"Fastai - progressive resizing","venue":null,"work_id":"37fd9832-3755-4fca-93b3-5c6571dd9078","year":2018},"citing_paper":{"arxiv_id":"1908.08986","last_updated":"2019-08-12T08:27:49Z","snapshot_observed_at":"2026-08-14T16:19:54.488219Z","submitted_at":"2019-08-12T08:27:49Z","title":"Mix & Match: training convnets with mixed image sizes for improved accuracy, speed and scale resiliency","version":1},"reference_index":2013,"source":"pdf_text","source_observed_at":"2026-08-14T13:59:54.918946Z"},"links":{"citing_paper":"/paper/1908.08986"},"observation_digest":"sha256:80451b3845aa2001926b72d2231f3311f14fce2ef9e45f4b42fea631b96c3614","observation_id":"195fd162-2dcf-4e45-9e67-13b5738fd7ba","resolution":{"observed_at":"2026-08-14T13:59:55.322027Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1902.05509","last_updated":"2019-04-03T08:07:29Z","snapshot_observed_at":"2026-08-14T17:16:16.614756Z","submitted_at":"2019-02-14T17:32:21Z","title":"MultiGrain: a unified image embedding for classes and instances","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1902.05509","snapshot_observed_at":"2026-08-14T13:59:54.876723Z","title":"Multigrain: a unified image embedding for classes and instances","venue":null,"work_id":null,"year":1902},"citing_paper":{"arxiv_id":"1908.08986","last_updated":"2019-08-12T08:27:49Z","snapshot_observed_at":"2026-08-14T16:19:54.488219Z","submitted_at":"2019-08-12T08:27:49Z","title":"Mix & Match: training convnets with mixed image sizes for improved accuracy, speed and scale resiliency","version":1},"reference_index":2014,"source":"pdf_text","source_observed_at":"2026-08-14T13:59:54.876723Z"},"links":{"cited_paper":"/paper/1902.05509","citing_paper":"/paper/1908.08986"},"observation_digest":"sha256:4351d6d7cdf0c92b32cde17f948a977d7e8e18e6bab848764ead6bba1bb724e8","observation_id":"1cf6d4e5-439a-421b-81f8-e190cebb4e05","resolution":{"observed_at":"2026-08-14T13:59:54.876723Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1710.10196","last_updated":"2018-02-26T15:33:34Z","snapshot_observed_at":"2026-08-15T04:29:56.166616Z","submitted_at":"2017-10-27T15:28:35Z","title":"Progressive Growing of GANs for Improved Quality, Stability, and Variation","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1710.10196","snapshot_observed_at":"2026-08-14T13:59:54.933686Z","title":"Progressive growing of gans for im- proved quality, stability, and variation","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"1908.08986","last_updated":"2019-08-12T08:27:49Z","snapshot_observed_at":"2026-08-14T16:19:54.488219Z","submitted_at":"2019-08-12T08:27:49Z","title":"Mix & Match: training convnets with mixed image sizes for improved accuracy, speed and scale resiliency","version":1},"reference_index":2015,"source":"pdf_text","source_observed_at":"2026-08-14T13:59:54.933686Z"},"links":{"cited_paper":"/paper/1710.10196","citing_paper":"/paper/1908.08986"},"observation_digest":"sha256:ad53991d3ceea6dc6bdf0a2e3cf83525b090127fa831818f52c0b151769ab94c","observation_id":"0ea53b7f-0b80-4f02-a9e2-e809d76a012a","resolution":{"observed_at":"2026-08-14T13:59:54.933686Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1411.6369","last_updated":"2014-11-24T07:28:21Z","snapshot_observed_at":"2026-08-14T23:10:30.802367Z","submitted_at":"2014-11-24T07:28:21Z","title":"Scale-Invariant Convolutional Neural Networks","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1411.6369","snapshot_observed_at":"2026-08-14T13:59:54.968858Z","title":"Scale-invariant con- volutional neural networks","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"1908.08986","last_updated":"2019-08-12T08:27:49Z","snapshot_observed_at":"2026-08-14T16:19:54.488219Z","submitted_at":"2019-08-12T08:27:49Z","title":"Mix & Match: training convnets with mixed image sizes for improved accuracy, speed and scale resiliency","version":1},"reference_index":2016,"source":"pdf_text","source_observed_at":"2026-08-14T13:59:54.968858Z"},"links":{"cited_paper":"/paper/1411.6369","citing_paper":"/paper/1908.08986"},"observation_digest":"sha256:16f3d97110b7e26c466d10ca359505476c26d3639cd2d4cbee208b4dd35d4a78","observation_id":"ec204c57-7cd9-4356-bb52-4ea7121f42bd","resolution":{"observed_at":"2026-08-14T13:59:54.968858Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1905.11286","last_updated":"2020-02-06T21:40:02Z","snapshot_observed_at":"2026-08-14T16:25:50.236639Z","submitted_at":"2019-05-27T15:12:50Z","title":"Stochastic Gradient Methods with Layer-wise Adaptive Moments for Training of Deep Networks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1905.11286","snapshot_observed_at":"2026-08-14T13:59:54.898794Z","title":"Stochastic gradient methods with layer- wise adaptive moments for training of deep networks","venue":null,"work_id":null,"year":1905},"citing_paper":{"arxiv_id":"1908.08986","last_updated":"2019-08-12T08:27:49Z","snapshot_observed_at":"2026-08-14T16:19:54.488219Z","submitted_at":"2019-08-12T08:27:49Z","title":"Mix & Match: training convnets with mixed image sizes for improved accuracy, speed and scale resiliency","version":1},"reference_index":2017,"source":"pdf_text","source_observed_at":"2026-08-14T13:59:54.898794Z"},"links":{"cited_paper":"/paper/1905.11286","citing_paper":"/paper/1908.08986"},"observation_digest":"sha256:10fd034c514b198d8bc5cdd1fae00eebaa6120934a5deb57213cde03a15fca5d","observation_id":"ac1e53a1-18b8-4d24-9c77-db61f82b6f8e","resolution":{"observed_at":"2026-08-14T13:59:54.898794Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1811.06965","last_updated":"2019-07-25T21:42:58Z","snapshot_observed_at":"2026-08-14T17:57:39.891981Z","submitted_at":"2018-11-16T18:43:28Z","title":"GPipe: Efficient Training of Giant Neural Networks using Pipeline Parallelism","version":5},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1811.06965","snapshot_observed_at":"2026-08-14T13:59:54.923849Z","title":"Gpipe: Efficient training of giant neural networks using pipeline parallelism.arXiv preprint arXiv:1811.06965,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"1908.08986","last_updated":"2019-08-12T08:27:49Z","snapshot_observed_at":"2026-08-14T16:19:54.488219Z","submitted_at":"2019-08-12T08:27:49Z","title":"Mix & Match: training convnets with mixed image sizes for improved accuracy, speed and scale resiliency","version":1},"reference_index":2018,"source":"pdf_text","source_observed_at":"2026-08-14T13:59:54.923849Z"},"links":{"cited_paper":"/paper/1811.06965","citing_paper":"/paper/1908.08986"},"observation_digest":"sha256:dc938fd74e999ed1df22cbeb52e9bbcfb2a305ec298c35b9b347d8fe098fc962","observation_id":"f5affd59-5f40-4632-addd-6187f75ff4a1","resolution":{"observed_at":"2026-08-14T13:59:54.923849Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1907.05550","last_updated":"2020-05-08T01:38:51Z","snapshot_observed_at":"2026-08-14T16:06:31.689600Z","submitted_at":"2019-07-12T02:17:12Z","title":"Faster Neural Network Training with Data Echoing","version":3},"cited_work":{"arxiv_id":"1907.05550","doi":null,"metadata_source":"pith","pith_arxiv_id":"1907.05550","snapshot_observed_at":"2026-08-14T13:59:55.237153Z","title":"Faster Neural Network Training with Data Echoing","venue":"cs.LG","work_id":"bb884721-da8a-4f9d-9ea1-6ad170cb2806","year":2019},"citing_paper":{"arxiv_id":"1908.08986","last_updated":"2019-08-12T08:27:49Z","snapshot_observed_at":"2026-08-14T16:19:54.488219Z","submitted_at":"2019-08-12T08:27:49Z","title":"Mix & Match: training convnets with mixed image sizes for improved accuracy, speed and scale resiliency","version":1},"reference_index":2019,"source":"pdf_text","source_observed_at":"2026-08-14T13:59:54.882361Z"},"links":{"cited_paper":"/paper/1907.05550","citing_paper":"/paper/1908.08986"},"observation_digest":"sha256:a010693a7617a7a77394cc597100632dc16b30d03ed5bce2090dd1cd7dbe6780","observation_id":"990fa8f0-be06-456d-a259-5a7c7def8aa9","resolution":{"observed_at":"2026-08-14T13:59:55.243925Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}}],"paper":{"arxiv_id":"1908.08986","last_updated":"2019-08-12T08:27:49Z","latest_version":1,"primary_category":"cs.CV","snapshot_observed_at":"2026-08-14T16:19:54.488219Z","submitted_at":"2019-08-12T08:27:49Z","title":"Mix & Match: training convnets with mixed image sizes for improved accuracy, speed and scale resiliency"},"reference_resolution":{"displayed":21,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":17,"verified_exact":1,"verified_fuzzy":3},"total_outbound_references":21},"refusal":"A citation records a reference. It does not transfer a finding from one paper to another.","schema":"pith.paper-citation-record.v1","standing_sources":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"thesis":"As of 17 August 2026, this Paper Citation Record lists 21 of 21 outbound references and 1 inbound Pith citation observation for arXiv:1908.08986."}