{"as_of":"2026-08-15T22:13:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:64efc282ed9de6085c1d8e9e9d3c39fd94af35cf44085b311ec65a1ac5458131","coverage":[{"denominator":76,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":76,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-14T13:54:34.916184Z","state":"measured"},{"denominator":76,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":76,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-15T06:32:42.880941+00:00","state":"measured"},{"denominator":0,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":0,"source":"paper_references, paper_reference_links","source_observed_at":null,"state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"cited_works","source_observed_at":null,"state":"measured"}],"external_citation_measurements":[],"inbound":[],"links":{"evidence":"/evidence","html":"/paper/1908.04207/citation-record","integrity":"/paper/1908.04207/integrity","json":"/paper/1908.04207/citation-record.json","paper":"/paper/1908.04207"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-14T13:54:34.620400Z","title":null,"venue":null,"work_id":null,"year":2015},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.620400Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:c2fe8f7370aa270f2e92540a5e901bf5005299d0656f0429d878515153ed0cda","observation_id":"d0003efc-7635-42a0-acf9-58fcff4b8c65","resolution":{"observed_at":"2026-08-14T13:54:34.620400Z","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":"2026-08-14T13:54:36.181553Z","title":null,"venue":null,"work_id":"e76d102e-cb4b-4f45-9380-c68421ae0e7a","year":2018},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.625214Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:8c440873ee5eafae456aaed350f5ae118c74dfe50f50fc5ad1567383df630fd5","observation_id":"d7f35542-7cf8-4f81-b0f4-d3b689752b19","resolution":{"observed_at":"2026-08-14T13:54:36.185244Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-14T13:54:36.170392Z","title":null,"venue":null,"work_id":"78294e8f-946b-4199-b87a-0b57a040e186","year":2018},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.629128Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:101df45b292b52f95713e8ce7026e3a6533eba194fa76885f7ef7877b2b74e2c","observation_id":"b3c21d8f-d333-4fc2-a051-f07de28119f9","resolution":{"observed_at":"2026-08-14T13:54:36.174451Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-14T13:54:36.159545Z","title":null,"venue":null,"work_id":"3ffc8a66-c41f-44a3-9328-e2cc538db49a","year":null},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.633460Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:196c6d0f9e33edaf6c1221c7d815d692537c70f45253c68617e8a8146eb86824","observation_id":"2084abb2-9266-44c2-89af-c375a9bc5aa2","resolution":{"observed_at":"2026-08-14T13:54:36.163399Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-14T13:54:36.147726Z","title":null,"venue":null,"work_id":"620e8d75-af9b-4662-94e2-54ea190aaedd","year":2017},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.641975Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:4a9b5b1b32a52f73b969cd3d21a4a9daf84e67145d5aa69890381b9efca27ef3","observation_id":"c39927c5-bbbf-4011-885d-3caffdd20b98","resolution":{"observed_at":"2026-08-14T13:54:36.151753Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-14T13:54:36.135548Z","title":null,"venue":null,"work_id":"f63957f8-27da-4488-82d7-780f8b80576e","year":2013},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.646030Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:404b608308333392cd93666395b263ce029629ba045a7c2b6278db27f9fad7fc","observation_id":"996cb1a7-9eff-49f0-82fa-7e622a468822","resolution":{"observed_at":"2026-08-14T13:54:36.139426Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1511.06432","last_updated":"2016-03-01T18:54:11Z","snapshot_observed_at":"2026-08-14T22:22:07.331593Z","submitted_at":"2015-11-19T22:46:13Z","title":"Delving Deeper into Convolutional Networks for Learning Video Representations","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1511.06432","snapshot_observed_at":"2026-08-14T13:54:34.650055Z","title":null,"venue":null,"work_id":null,"year":2015},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.650055Z"},"links":{"cited_paper":"/paper/1511.06432","citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:c59ec94bf5665bf05b8e7b25f23d035582f118bc6f727a916036c15dcad25e6f","observation_id":"11c36705-3688-4b3a-b097-620d54bbd69d","resolution":{"observed_at":"2026-08-14T13:54:34.650055Z","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":"2026-08-14T13:54:36.123802Z","title":null,"venue":null,"work_id":"4af2e69b-f366-45ee-9814-e2c54c693391","year":2018},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.654939Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:56bc66712e31859582f7b7341a975f97d84cd01a4b25f057b3ef3a42a3bc689a","observation_id":"da95263c-e59e-46ae-8df3-939f7fcec931","resolution":{"observed_at":"2026-08-14T13:54:36.127479Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-14T13:54:36.111649Z","title":"Ben-Nun, M","venue":null,"work_id":"a34aa742-00b7-4c87-964a-98a5f08b3ed1","year":null},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.659051Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:edc1e79ea67eb2776c83332df9965fd012de29beaa740e9a8f2692344a760d8b","observation_id":"f6d53a3a-ea35-4ea3-8ead-21371d213034","resolution":{"observed_at":"2026-08-14T13:54:36.115948Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1802.09941","last_updated":"2018-09-15T08:36:28Z","snapshot_observed_at":"2026-08-15T04:18:46.141162Z","submitted_at":"2018-02-26T08:47:34Z","title":"Demystifying Parallel and Distributed Deep Learning: An In-Depth Concurrency Analysis","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1802.09941","snapshot_observed_at":"2026-08-14T13:54:34.667415Z","title":"Ben-Nun and T","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.667415Z"},"links":{"cited_paper":"/paper/1802.09941","citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:31f7520913f36247059be3f9a489625b66bac93d9dd48629a932bd4c381a9877","observation_id":"4c76b4c7-3f79-4eb2-a01a-f4936f6c9b8e","resolution":{"observed_at":"2026-08-14T13:54:34.667415Z","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":"2026-08-14T13:54:36.100979Z","title":null,"venue":null,"work_id":"4b1d1939-8aee-469e-94ad-00f509fcef87","year":2014},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.671995Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:b0664783cc53a2221af0e6bd42a59b5508710b193b24d01e3de04be2355fe06f","observation_id":"94fe8eec-684f-4c90-b98e-78a7adb11946","resolution":{"observed_at":"2026-08-14T13:54:36.104443Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1803.05880","last_updated":"2018-03-15T17:32:16Z","snapshot_observed_at":"2026-08-14T19:35:44.874756Z","submitted_at":"2018-03-15T17:32:16Z","title":"GossipGraD: Scalable Deep Learning using Gossip Communication based Asynchronous Gradient Descent","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1803.05880","snapshot_observed_at":"2026-08-14T13:54:34.675971Z","title":null,"venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.675971Z"},"links":{"cited_paper":"/paper/1803.05880","citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:42259d9e4cfd9ee37c61649e8ac0dbf90418adb6c32c08bfe9ad6d4adb541870","observation_id":"312d0c1e-5ef6-4fc4-b6c2-09fc0643d29b","resolution":{"observed_at":"2026-08-14T13:54:34.675971Z","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":"2026-08-14T13:54:36.089900Z","title":null,"venue":null,"work_id":"edac38ec-c6a5-45b6-a252-0854d5dc2560","year":2012},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.679961Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:4c4288b311d287b238c9dba8a6260c73897efda0e3c0dcc5deb444a76589d635","observation_id":"7895aa13-295f-48d0-a64e-5a20f0f447e8","resolution":{"observed_at":"2026-08-14T13:54:36.093508Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-14T13:54:34.683757Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.683757Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:9a1cb8cdfa6015817242fafa3c0e17d618a0156df612e0704df717ff396c59b9","observation_id":"bd931a11-c5fd-410b-a02f-b685098f4e33","resolution":{"observed_at":"2026-08-14T13:54:34.683757Z","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":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-14T13:54:34.691983Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.691983Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:3225fe6bcba6d53c89d780fec656be98de1a47e9505a78822145355762bec8dd","observation_id":"78a7a49f-ce7b-4233-88d4-bb0cc37e575e","resolution":{"observed_at":"2026-08-14T13:54:34.691983Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"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":"2026-08-14T13:54:34.700312Z","title":null,"venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.700312Z"},"links":{"cited_paper":"/paper/1708.04552","citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:ec5808cacb3b326ad4bb1585c01a96879d18e4de06d5fe40f0b6ebf961b1ec05","observation_id":"c27384cf-1112-4ef5-9a31-767924671c6a","resolution":{"observed_at":"2026-08-14T13:54:34.700312Z","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":"2026-08-14T13:54:36.052988Z","title":null,"venue":null,"work_id":"4cc7995d-157b-4521-bc36-bf9dbdeb9639","year":2015},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.704369Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:e5483e43d2c1728a95cbdd6c7a83efa2f2ea1e02c6bffeeca44aec3339130306","observation_id":"c97cfb5c-7b0f-4783-a589-70bffb1e8f54","resolution":{"observed_at":"2026-08-14T13:54:36.056803Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-14T13:54:36.040789Z","title":null,"venue":null,"work_id":"a21e0413-4212-4706-ab78-27d8a58983df","year":null},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.707792Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:55484363723efebc79d042bdc685f13d393908b88bbef7d27923f996ea79babe","observation_id":"ec211836-5b95-40f7-82a6-f38e933bcb31","resolution":{"observed_at":"2026-08-14T13:54:36.045251Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1810.04805","last_updated":"2019-05-24T20:37:26Z","snapshot_observed_at":"2026-08-14T18:16:28.847993Z","submitted_at":"2018-10-11T00:50:01Z","title":"BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1810.04805","snapshot_observed_at":"2026-08-14T13:54:34.696203Z","title":"CoRR abs/1810.04805 (2018)","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.696203Z"},"links":{"cited_paper":"/paper/1810.04805","citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:1ab8eb329d7713129b964f188be818cdd60b3ce0e27c6233d3041bf5f4893681","observation_id":"205e657e-c11e-4465-b52d-333d3f1a9e53","resolution":{"observed_at":"2026-08-14T13:54:34.696203Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1509.04210","last_updated":"2016-12-05T21:26:38Z","snapshot_observed_at":"2026-08-14T22:32:28.043611Z","submitted_at":"2015-09-14T17:14:52Z","title":"Model Accuracy and Runtime Tradeoff in Distributed Deep Learning:A Systematic Study","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1509.04210","snapshot_observed_at":"2026-08-14T13:54:34.719597Z","title":null,"venue":null,"work_id":null,"year":2015},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.719597Z"},"links":{"cited_paper":"/paper/1509.04210","citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:c762e868e23676f94109c7edf3170eb4faa1e5410f5332b6a499e5426fc87c29","observation_id":"d910b9b4-bd6a-463d-9fbd-1f7f1b8aad5e","resolution":{"observed_at":"2026-08-14T13:54:34.719597Z","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":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-14T13:54:34.724029Z","title":null,"venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.724029Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:168098cd8bcbe3d6394ee3be422b1b1ae7968004b1806eab7fb58e8f56bfe8f7","observation_id":"374b8110-172f-402b-925b-e1808be9d22b","resolution":{"observed_at":"2026-08-14T13:54:34.724029Z","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":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-14T13:54:34.727704Z","title":"Gibbons, Garth A","venue":null,"work_id":null,"year":2013},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.727704Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:f4e05c95cc5b29a582646573d2528a291cf992c09202b81e325d86f45aa035cb","observation_id":"28e27786-0dc4-4e26-b533-92233267a6f4","resolution":{"observed_at":"2026-08-14T13:54:34.727704Z","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":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-14T13:54:34.731168Z","title":null,"venue":null,"work_id":null,"year":1997},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.731168Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:9f64c5a856c69b1d464e19e84af8a0475345e039a4f131f70b0d1de7a671ceab","observation_id":"4307757d-61ea-4105-a37f-ee932ec4c747","resolution":{"observed_at":"2026-08-14T13:54:34.731168Z","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":"2026-08-14T13:54:36.028659Z","title":null,"venue":null,"work_id":"13f80fd8-d9e3-4764-8e2d-a5c1de0daea9","year":2017},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.715794Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:9caf8645e038525ecdf2354fbd8d33a7c2c96da8186fd7870a1000ac0152bce0","observation_id":"f3fdfdd3-9b3b-4f15-8ab0-0b588c253e80","resolution":{"observed_at":"2026-08-14T13:54:36.032694Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-14T13:54:35.990068Z","title":null,"venue":null,"work_id":"2bc17a2b-97f0-4a82-aabc-987d339b3a3b","year":2007},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.739001Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:383007ebb535929d5707ad31c14c8520a3a3d647bebc28521baf99382b006623","observation_id":"4cfa9348-b143-405b-8783-dc91b41e9be0","resolution":{"observed_at":"2026-08-14T13:54:35.994145Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-14T13:54:35.978110Z","title":"Hoefler and D","venue":null,"work_id":"b63a63b4-eb5f-4ea6-b056-98bf9dabbdfe","year":2014},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.743121Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:2be7c54ddc01fcbd885939363782576cdf482c240d18a51a27014e8ee1e5f20c","observation_id":"857dcb7f-f52a-4f64-b8a7-d8ae5829a472","resolution":{"observed_at":"2026-08-14T13:54:35.982242Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-14T13:54:35.964430Z","title":"Hoefler, T","venue":null,"work_id":"3907a98e-172c-4c9c-b402-34b9ad8696ff","year":2009},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.746703Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:e86511c3a687edb46485da987eac00fc2c3e60ee71a8b2bf2b84d7e92b319d0d","observation_id":"376d0997-81e7-4ee6-9832-09096a3de59f","resolution":{"observed_at":"2026-08-14T13:54:35.968985Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-14T13:54:35.951308Z","title":"Hoefler, T","venue":null,"work_id":"fe2a17a4-b5ae-4ef6-924c-994dfb191db3","year":2010},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.750358Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:d4d848eaefd3eb9068addba901ab58d9234348b920fa4062a9ee06f8ea639e5c","observation_id":"7e24dea2-5cdb-47e4-bf7c-37c9b48c84c3","resolution":{"observed_at":"2026-08-14T13:54:35.955373Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-14T13:54:36.001860Z","title":null,"venue":null,"work_id":"127fb325-efc4-418c-bb45-9399435bf46f","year":2017},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.735198Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:10dfd2768c79aa10bb3b028b182d884739ac921900b3933f962b590aedf624bb","observation_id":"b557eeef-f9d5-4b03-9b66-300db51ab9bd","resolution":{"observed_at":"2026-08-14T13:54:36.006013Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-14T13:54:34.757646Z","title":"Huang, Z","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.757646Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:85220db612b192936573366eb3ed820b1ab91a875bfa914969f2678e811d2c23","observation_id":"9ebc1ef6-3c8d-4cc3-994d-1689f3d84158","resolution":{"observed_at":"2026-08-14T13:54:34.757646Z","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":"2026-08-14T13:54:35.938551Z","title":null,"venue":null,"work_id":"4d88c6d1-97cb-45e0-b241-3fc1d8fbf555","year":2011},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.761873Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:2cb81535c3b63f8ec2d2dd0837d426ca5afcceb0f06f880f3222dde4b1b260da","observation_id":"1caa638d-23a6-46fb-8981-1706b32e49b2","resolution":{"observed_at":"2026-08-14T13:54:35.943086Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-14T13:54:35.926470Z","title":null,"venue":null,"work_id":"c5891f80-ea5a-456f-8015-979d4897193f","year":2010},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.765570Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:667e9449912911671982ec6fb9f8a95cb138ca4ae8d116df4c2e43378faf4df0","observation_id":"9b9dc776-0ac0-4452-9eff-83898e1b7ae1","resolution":{"observed_at":"2026-08-14T13:54:35.930419Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-14T13:54:35.915062Z","title":null,"venue":null,"work_id":"6b0ec3de-7609-497d-bf5a-5654cf62d0d3","year":2019},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.769760Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:2f8abefb0fbc0c0835c321fc69453cfdf82095189fbd4901722962d7562cbe5e","observation_id":"90bccd2e-45b8-4157-9217-2fef99459160","resolution":{"observed_at":"2026-08-14T13:54:35.918798Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"4630.31546","doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-14T13:54:35.383388Z","title":"Ganger, Phillip B","venue":null,"work_id":"1320a4f3-c6fb-4f76-91d2-3256b21d6dc3","year":2017},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.753977Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:bce8f05082c9f65bcdfe102f4fea3848a97aaf71c66cc8c2a49f6ec0513d61f7","observation_id":"e41cc853-3bda-4958-bd3a-7bb731be33ff","resolution":{"observed_at":"2026-08-14T13:54:35.389473Z","resolver_source":"raw_fallback","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-14T13:54:34.777667Z","title":null,"venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.777667Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:823d7f99f3eb3b133cd604aac088633e5e73806486f766e1e3f5ecf0fa3cb1b4","observation_id":"bd9c6f6d-92d3-4100-8460-43aabc356e89","resolution":{"observed_at":"2026-08-14T13:54:34.777667Z","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":"2026-08-14T13:54:35.902552Z","title":"LeCun, Y","venue":null,"work_id":"7e8c5c64-1e02-4052-9b64-c49e35578e9e","year":2015},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.781429Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:20935e6717333d25735d6d42bfd3577e106129d804ca1241be0dc00cf396cc1e","observation_id":"504ffaf9-f40b-4597-aab7-c099c4c5ae9d","resolution":{"observed_at":"2026-08-14T13:54:35.906813Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-14T13:54:34.785165Z","title":null,"venue":null,"work_id":null,"year":1998},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.785165Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:8d072f59c59efc51a6b85bb70b58e0a906e50f254303d490d013a74129ce8aea","observation_id":"e1a21e9c-7252-4335-8c4b-9d36b5da772e","resolution":{"observed_at":"2026-08-14T13:54:34.785165Z","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":"2026-08-14T13:54:35.884670Z","title":"Andersen, Jun Woo Park, Alexander J","venue":null,"work_id":"b5760b5f-1287-41b5-8e42-9d1063a1104f","year":2014},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.788816Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:515db45ee86df72c01b3688737d6bf11636fbe3890caeac64f9016032d106e52","observation_id":"5812f867-1d1a-4ffa-a2de-ed81e26231ff","resolution":{"observed_at":"2026-08-14T13:54:35.888221Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1611.04581","last_updated":"2016-11-14T20:59:54Z","snapshot_observed_at":"2026-08-14T21:30:28.408283Z","submitted_at":"2016-11-14T20:59:54Z","title":"How to scale distributed deep learning?","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1611.04581","snapshot_observed_at":"2026-08-14T13:54:34.773511Z","title":"Jin, Qiaochu Yuan, Forrest N","venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.773511Z"},"links":{"cited_paper":"/paper/1611.04581","citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:f4eb87c7a8c90405812c3955e9e5a6e463a73a84594f5bf6371e3730fbe2b48d","observation_id":"0ae32da0-7e30-40ee-82b3-13cea8b66514","resolution":{"observed_at":"2026-08-14T13:54:34.773511Z","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":"2026-08-14T13:54:35.873510Z","title":null,"venue":null,"work_id":"6db0044e-d1b9-4fff-8780-d48c2fff6472","year":2018},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.796633Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:5cb7341aea09b15f76dc84fc05db4095514014f3ac429ab0c1dc18416c68e998","observation_id":"c8439ce7-1b8b-4ac1-b220-7f7d568a7504","resolution":{"observed_at":"2026-08-14T13:54:35.877070Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-14T13:54:34.800189Z","title":"Pennycook, Kristyn Maschhoff, Jason Sewall, Nalini Kumar, Shirley Ho, Michael F","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.800189Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:c9b6e44cb79f35cfd472287e91be49d896714a2d45f545db13d2a83afb59bdde","observation_id":"bcd56820-aa9f-48b3-949c-83eee763e2ff","resolution":{"observed_at":"2026-08-14T13:54:34.800189Z","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":"2026-08-14T13:54:35.863046Z","title":null,"venue":null,"work_id":"13e0324e-1744-4e90-81c8-8958a2a1dd3f","year":2015},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.803755Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:6fba000d4804a7c9a93171caa74da94f65ba235f81d6c8e72ffaac9c2da134ef","observation_id":"92ce3947-9509-49f1-a1a5-a6379dfa96da","resolution":{"observed_at":"2026-08-14T13:54:35.866619Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-14T13:54:34.807747Z","title":null,"venue":null,"work_id":null,"year":2009},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.807747Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:9ab3471cf580af9f9c4b9c51d2912b6df01bfbb256375844188b4ad73dc54ef1","observation_id":"8c6a4e8a-2bcf-4f3f-bafb-a8854afe90b7","resolution":{"observed_at":"2026-08-14T13:54:34.807747Z","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":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-14T13:54:34.793076Z","title":null,"venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":44,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.793076Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:7545ebb5da96e7c78b8744f5e555e9fa7e3ca820861c22e99f3b4a48efdbfca5","observation_id":"e0352d87-1f77-4c5c-ac0c-edab6c3720e3","resolution":{"observed_at":"2026-08-14T13:54:34.793076Z","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":"2026-08-14T13:54:35.838883Z","title":null,"venue":null,"work_id":"6f6b875d-1f3a-416a-b004-5dce48965c78","year":2018},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":45,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.815289Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:1773b82c938be322fe81528b9bf6a73703d9544b0f8a8556178fdc4eb3ceee22","observation_id":"08742fa5-8aa0-4010-bd14-4ba2cb729e32","resolution":{"observed_at":"2026-08-14T13:54:35.843425Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-14T13:54:35.826691Z","title":"Recht, C","venue":null,"work_id":"b5eb7ef7-4109-411f-93fa-b8356c5e3638","year":2011},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":46,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.819018Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:4005b1947e47eb6e6ab0f16b5392033884cbf35eb37cf2dd9969b8fc6dd48477","observation_id":"dd0d74a7-f9c3-4cee-b6be-4642c3e1726e","resolution":{"observed_at":"2026-08-14T13:54:35.830344Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1802.08021","last_updated":"2019-08-16T07:46:14Z","snapshot_observed_at":"2026-08-14T19:43:20.964402Z","submitted_at":"2018-02-22T13:02:53Z","title":"SparCML: High-Performance Sparse Communication for Machine Learning","version":3},"cited_work":{"arxiv_id":"1802.08021","doi":null,"metadata_source":"pith","pith_arxiv_id":"1802.08021","snapshot_observed_at":"2026-08-14T13:54:35.098153Z","title":"SparCML: High-Performance Sparse Communication for Machine Learning","venue":"cs.DC","work_id":"98148bea-5de4-4a62-90b3-16ac762c09a9","year":2018},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":47,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.822558Z"},"links":{"cited_paper":"/paper/1802.08021","citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:05b6363c4f53741c751195b9747579c226e7eea8040b31cf12572df511d25e32","observation_id":"fac1964e-baed-4619-9df1-d7cffc6ad4ee","resolution":{"observed_at":"2026-08-14T13:54:35.103747Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-14T13:54:35.815868Z","title":null,"venue":null,"work_id":"a0e43d16-70a8-45b9-a129-d4dcaf714477","year":1951},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":48,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.826476Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:79c4b2685096c7f68285c1633935afd48d76e169c27ec063c5b856aeb9ebda8f","observation_id":"80357582-2617-44ed-b7d9-24064fbff5ad","resolution":{"observed_at":"2026-08-14T13:54:35.819622Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-14T13:54:35.851725Z","title":null,"venue":null,"work_id":"5f434f7b-84b2-416d-8eb3-10b837aaa0ea","year":2004},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":49,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.811636Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:e36237d9a7cd5b73548f33ec24a58c1b27b03fb74ef1aaea1f9e01e25be0127e","observation_id":"28ad2753-bf30-4ecb-b3f2-ffa2f8e4d80a","resolution":{"observed_at":"2026-08-14T13:54:35.855619Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-14T13:54:35.793823Z","title":null,"venue":null,"work_id":"efe3ec24-ef40-4177-8ef9-39dfdae1858f","year":2014},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":50,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.834036Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:dcdb44e5e99e872be142e48271cee3d2df283f4312bb2edb3e381f4ad930aa50","observation_id":"956f4120-9fdc-4e54-b824-3d8a691ba3ee","resolution":{"observed_at":"2026-08-14T13:54:35.797951Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1802.05799","last_updated":"2018-02-21T04:30:30Z","snapshot_observed_at":"2026-08-14T19:45:22.282120Z","submitted_at":"2018-02-15T23:36:51Z","title":"Horovod: fast and easy distributed deep learning in TensorFlow","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1802.05799","snapshot_observed_at":"2026-08-14T13:54:34.838921Z","title":null,"venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":51,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.838921Z"},"links":{"cited_paper":"/paper/1802.05799","citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:299db37e506813e5e1b8224d3fd58431d265653f3ddd190a0664b2a7d7fb8915","observation_id":"361aea5d-a1bb-4f04-aba5-dd6e8f853bcc","resolution":{"observed_at":"2026-08-14T13:54:34.838921Z","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":"2026-08-14T13:54:34.843055Z","title":null,"venue":null,"work_id":null,"year":2014},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":52,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.843055Z"},"links":{"cited_paper":"/paper/1409.1556","citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:0f2b834269e551c2112e9f4eb1739e22b4f973b0a17a283eab84be245f93b008","observation_id":"d7a32e3a-f78e-4ab4-af82-b994cada1925","resolution":{"observed_at":"2026-08-14T13:54:34.843055Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1212.0402","last_updated":"2012-12-03T14:45:31Z","snapshot_observed_at":"2026-08-15T13:34:03.745436Z","submitted_at":"2012-12-03T14:45:31Z","title":"UCF101: A Dataset of 101 Human Actions Classes From Videos in The Wild","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1212.0402","snapshot_observed_at":"2026-08-14T13:54:34.847478Z","title":null,"venue":null,"work_id":null,"year":2012},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":53,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.847478Z"},"links":{"cited_paper":"/paper/1212.0402","citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:9105a6147c2650cd3d66f7c276e2b991881028baf39140ad40aaec89d25685f9","observation_id":"eb9fd678-f471-4be6-9313-c6973dd05c11","resolution":{"observed_at":"2026-08-14T13:54:34.847478Z","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":"2026-08-14T13:54:35.805354Z","title":null,"venue":null,"work_id":"0af2000c-8edb-4e60-8f7c-944adfa0018d","year":2010},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":54,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.830205Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:aec3ce67942a3388a099b8643c6082bbc4d3565a761f8f237f8cc7fd62855088","observation_id":"f9802b21-ce77-471e-896c-9850ec3c0189","resolution":{"observed_at":"2026-08-14T13:54:35.809021Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-14T13:54:34.855548Z","title":null,"venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":55,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.855548Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:002bebd3ff1aeeb9cf4288dd50cac97ae31218218d1cb98e09de3f162f80ac46","observation_id":"231761cc-026c-46fb-97b6-1ba70e901b56","resolution":{"observed_at":"2026-08-14T13:54:34.855548Z","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":"2026-08-14T13:54:35.765050Z","title":null,"venue":null,"work_id":"21918946-5e20-4a67-bbba-458f3e435695","year":2005},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":56,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.859066Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:74cc16da35460b52e37b5b8575ab9ba5c00f2f3715c1af88051159c62715746e","observation_id":"3462575c-bfde-4155-866a-fb0f2386f1cc","resolution":{"observed_at":"2026-08-14T13:54:35.769022Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-14T13:54:35.754309Z","title":null,"venue":null,"work_id":"30f4e5af-51db-4f46-b329-469cb52e7ea2","year":2017},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":57,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.863181Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:f1e034a85f359e6598ee95afc90d3dc5ff84df89faa3356e2be298933f6f8c33","observation_id":"28786b36-b4f1-4c1e-ae0b-e27f2ad7810a","resolution":{"observed_at":"2026-08-14T13:54:35.758169Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"0948.30210","doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-14T13:54:35.048019Z","title":null,"venue":null,"work_id":"037cc05d-f839-4ce3-89a4-70b86d4ef3b0","year":2016},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":58,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.867394Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:07744f0e90334927c3338b4b1f12af1b724ceaba651ce421c6e007ee84ba0a60","observation_id":"975f483c-07b2-48ca-bb45-818ff0997500","resolution":{"observed_at":"2026-08-14T13:54:35.054403Z","resolver_source":"raw_fallback","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-14T13:54:35.782437Z","title":null,"venue":null,"work_id":"a683fbdd-9751-4402-ac70-2c4a86f20659","year":2015},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":59,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.851647Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:44c7987e9b2cfda6205a4c54eea939e6f5b0ad54988e9c4077f7982f16b690d2","observation_id":"98f4dca1-ed85-4d91-9df2-6fd72eff23e2","resolution":{"observed_at":"2026-08-14T13:54:35.785911Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-14T13:54:35.731342Z","title":null,"venue":null,"work_id":"bf47e94f-449f-4f17-8d4e-250127cc7879","year":2015},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":60,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.874605Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:b1e7bd1e05fcf102897ffa6604daf338411aa7c6d1a56eabda0447a393a47abe","observation_id":"7e8f1b7d-98b2-437c-9166-4a3ffbc92b5a","resolution":{"observed_at":"2026-08-14T13:54:35.735670Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-14T13:54:35.720099Z","title":null,"venue":null,"work_id":"f6879a6f-149a-44c8-8441-9baa259b0dff","year":2015},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":61,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.878121Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:4561fa9af502b4f507a6942f49fb77e13cc21995f62af292cfa0bc9995a17dc3","observation_id":"072206d2-5fcf-41c1-8bb9-891d93fd4caa","resolution":{"observed_at":"2026-08-14T13:54:35.723867Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1511.05950","last_updated":"2016-04-05T06:21:03Z","snapshot_observed_at":"2026-08-14T22:22:20.430156Z","submitted_at":"2015-11-18T20:53:33Z","title":"Staleness-aware Async-SGD for Distributed Deep Learning","version":5},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1511.05950","snapshot_observed_at":"2026-08-14T13:54:34.882444Z","title":null,"venue":null,"work_id":null,"year":2015},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":62,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.882444Z"},"links":{"cited_paper":"/paper/1511.05950","citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:968256bf1055b3043a86e54fc9dd1d66fce08c935afbb3e7995bef3c587b04d9","observation_id":"cdbeebcc-81c0-43b2-baa7-e964935e4427","resolution":{"observed_at":"2026-08-14T13:54:34.882444Z","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":"2026-08-14T13:54:35.742863Z","title":null,"venue":null,"work_id":"f3280f1f-b18e-4722-9561-ffc21f0234b8","year":2018},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":64,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.871063Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:f0a88439ca55a2a2555a63eb270096e2341f08b39fd26c92e680c6b06f4ca729","observation_id":"5f836607-d59c-4de8-9fbc-ed0f12e6a305","resolution":{"observed_at":"2026-08-14T13:54:35.747213Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-14T13:54:35.709546Z","title":"Extract the artifact using unzip","venue":null,"work_id":"3a8237ff-a00b-4c5c-8115-2d998cb0ddd6","year":null},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":68,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.886416Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:fdd252915ca8f515f8bee4165eeef8e650d1f110ebb37f3a84ac52fd6dd74bf8","observation_id":"9d434c47-b687-4372-9c81-74dd07d25f38","resolution":{"observed_at":"2026-08-14T13:54:35.713055Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-14T13:54:35.697829Z","title":"> cd $WORK/eager-SGD-artifact/eager-SGD > pip install -r requirements.txt","venue":null,"work_id":"d130dee2-0465-4976-86a3-39cba77511a4","year":null},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":69,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.890381Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:6d29df8e4c99362f0522e021a14cea7f6cb7230f1d47788d8068a04861ebb65a","observation_id":"1c7d4e9d-f759-4f3e-a2e8-85a8591f7e08","resolution":{"observed_at":"2026-08-14T13:54:35.701984Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-14T13:54:35.686400Z","title":"> cd $WORK/eager-SGD-artifact/eager-SGD/fflib2/lib > cmake","venue":null,"work_id":"36c5b37d-5196-4c45-8613-de823fe1451a","year":null},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":70,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.894012Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:2d5ff2e0ef69bb695f6e3da8f3aaf24414239aeb3d561c6cc95cb884d2fd2840","observation_id":"ba72363c-4ca1-479e-b062-53f7debf3941","resolution":{"observed_at":"2026-08-14T13:54:35.690582Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-14T13:54:35.674285Z","title":null,"venue":null,"work_id":"29bfe00d-80b1-400f-bab4-c50db2f85629","year":null},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":71,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.897651Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:f88683f3d84be48f88179848f505a368fa3bac63635e346b2a93942c19617f48","observation_id":"5d04c620-ff0f-40a2-8522-ef188da6b919","resolution":{"observed_at":"2026-08-14T13:54:35.678495Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-14T13:54:35.662045Z","title":"<\" means","venue":null,"work_id":"995a0f0f-bba0-4cf6-8f50-e452436975be","year":null},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":72,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.901713Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:3add7fce0537deed300995f46f5a2d883752462054c62f5c3e6d6a2731b90c86","observation_id":"afd0a148-cbe2-4a81-bea3-be8e0a762005","resolution":{"observed_at":"2026-08-14T13:54:35.666228Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-14T13:54:35.650704Z","title":null,"venue":null,"work_id":"cdeba060-6cc5-4b57-9390-310f76acdd47","year":null},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":73,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.905236Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:cd32b2d9367a186e35d13466ee547a6326f79f7eb96c5e0a6bce1dfddf8c1c02","observation_id":"b3784f23-7898-4099-b239-30fbe751894a","resolution":{"observed_at":"2026-08-14T13:54:35.654355Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-14T13:54:35.639217Z","title":null,"venue":null,"work_id":"ffa4667b-3238-47da-8011-57cd17ba4583","year":null},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":74,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.908915Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:d5c0291fbd13e001e7fa4d387c354d1c14f6e91b1dee6075d4ae57e767b4b1eb","observation_id":"db07171a-64e0-45b5-add7-8054f1baa8c8","resolution":{"observed_at":"2026-08-14T13:54:35.643189Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-14T13:54:35.626829Z","title":"<\" means","venue":null,"work_id":"e9744839-24bd-4ef3-9234-266bed0bfeba","year":null},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":75,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.912662Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:7959122856af725ca0c259634b23dd418885fe81952b107fc90c7a4c1f7136ea","observation_id":"d1b38445-4ee4-4df8-be6e-ea0af010e55c","resolution":{"observed_at":"2026-08-14T13:54:35.630930Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-14T13:54:35.615598Z","title":"<\" means","venue":null,"work_id":"3f5967e2-daf3-4f72-a255-a378adcb8abb","year":null},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":76,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.916184Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:6fc6ab9446a3de9b1d8241220ad7e1d095b1696b9c434966a3c659888dcf1b2f","observation_id":"7dde7bbe-6a07-4321-bc57-ff6edf3fa637","resolution":{"observed_at":"2026-08-14T13:54:35.619419Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-14T13:54:36.070949Z","title":"In Pro- ceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)","venue":null,"work_id":"a2254adb-9d7b-47b9-8213-a1fc7c983d11","year":2009},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":2009,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.687674Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:4a9097cd1c823b4ec0686656ac04e91c7d45fc67a7e2bcec4e2b292dbb3b2d5a","observation_id":"3b231cda-28dd-49ba-bee9-94e689565f2b","resolution":{"observed_at":"2026-08-14T13:54:36.075252Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1411.4389","last_updated":"2016-05-31T22:57:33Z","snapshot_observed_at":"2026-08-14T23:11:24.773671Z","submitted_at":"2014-11-17T08:25:17Z","title":"Long-term Recurrent Convolutional Networks for Visual Recognition and Description","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1411.4389","snapshot_observed_at":"2026-08-14T13:54:34.711661Z","title":"CoRR abs/1411.4389 (2014)","venue":null,"work_id":null,"year":2014},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":2014,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.711661Z"},"links":{"cited_paper":"/paper/1411.4389","citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:4898c87cea132458d80deec266b3f7d679651a64231ea01d98f50fe936fe6196","observation_id":"17f3ce56-c852-4ffc-96d4-734efbf9e090","resolution":{"observed_at":"2026-08-14T13:54:34.711661Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1811.10792","last_updated":"2019-05-14T19:59:00Z","snapshot_observed_at":"2026-08-14T17:52:54.241865Z","submitted_at":"2018-11-27T03:47:26Z","title":"Stochastic Gradient Push for Distributed Deep Learning","version":3},"cited_work":{"arxiv_id":"1811.10792","doi":null,"metadata_source":"pith","pith_arxiv_id":"1811.10792","snapshot_observed_at":"2026-08-14T13:54:35.601717Z","title":"Stochastic Gradient Push for Distributed Deep Learning","venue":"cs.LG","work_id":"52059521-cf1c-40fe-9011-84d6ebebaa35","year":2018},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":2018,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.637637Z"},"links":{"cited_paper":"/paper/1811.10792","citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:5e4c216b313a99b18749f89570854ea5b990712935596ad28f9716e12e393b56","observation_id":"651b2244-d216-4c23-99f9-5b76c1ede3d4","resolution":{"observed_at":"2026-08-14T13:54:35.606664Z","resolver_source":"local_arxiv","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-14T13:54:34.663467Z","title":"In 2019 IEEE International Parallel and Distributed Processing Symposium (IPDPS)","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations","version":5},"reference_index":2019,"source":"pdf_text","source_observed_at":"2026-08-14T13:54:34.663467Z"},"links":{"citing_paper":"/paper/1908.04207"},"observation_digest":"sha256:bf892806509d231dd32f7104a51d4a46f5ef2f3b6a40cd2fed24eb654077bf56","observation_id":"ff9f93ce-0417-4900-a6bd-28d088f9ae2b","resolution":{"observed_at":"2026-08-14T13:54:34.663467Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"paper":{"arxiv_id":"1908.04207","last_updated":"2025-08-21T07:39:24Z","latest_version":5,"primary_category":"cs.DC","snapshot_observed_at":"2026-08-15T08:39:23.320015Z","submitted_at":"2019-08-12T15:37:51Z","title":"Taming Unbalanced Training Workloads in Deep Learning with Partial Collective Operations"},"reference_resolution":{"displayed":76,"state_counts":{"malformed_identifier":0,"metadata_mismatch":3,"parse_uncertain":0,"unresolved":58,"verified_exact":1,"verified_fuzzy":14},"total_outbound_references":76},"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":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"thesis":"As of 15 August 2026, this Paper Citation Record lists 76 of 76 outbound references and 0 inbound Pith citation observations for arXiv:1908.04207."}