{"as_of":"2026-08-19T07:58:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:cc2ff100d03cfb908cfbff96c5c06785bce4ee45a73caabac1ffa2a436e02cfb","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":"2026-08-08T14:14:35.448676Z","state":"measured"},{"denominator":80,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":80,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-19T06:32:44.657259+00:00","state":"measured"},{"denominator":4,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":4,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-07T04:57:52.568195Z","state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"arxiv_reference","source_observed_at":"2026-07-01T21:36:15.542964Z","state":"measured"}],"external_citation_measurements":[],"inbound":[{"citation":{"cited_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2502.06982","snapshot_observed_at":"2026-08-07T04:57:52.568195Z","title":"Machine learning fleet efficiency: Analyzing and optimizing large-scale google tpu systems with ml productivity goodput.arXiv preprint arXiv:2502.06982, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2506.09275","last_updated":"2025-06-10T22:25:29Z","snapshot_observed_at":"2026-08-14T16:42:28.054150Z","submitted_at":"2025-06-10T22:25:29Z","title":"A Survey of End-to-End Modeling for Distributed DNN Training: Workloads, Simulators, and TCO","version":1},"reference_index":129,"source":"pdf_text","source_observed_at":"2026-08-07T04:57:52.568195Z"},"links":{"cited_paper":"/paper/2502.06982","citing_paper":"/paper/2506.09275"},"observation_digest":"sha256:0e64fa32f0743a455cc6abd01b10c8cce5b874a091c952442513cad58c825085","observation_id":"41d4bd1b-5ccd-4729-80ed-24c41b4d0ce2","resolution":{"observed_at":"2026-08-07T04:57:52.568195Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"cited_work":{"arxiv_id":"2502.06982","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2502.06982","snapshot_observed_at":"2026-07-01T21:36:15.542964Z","title":"Ma- chine learning fleet efficiency: analyzing and optimizing large-scale Google TPU systems with ML productivity goodput","venue":null,"work_id":"0a235eaf-8a18-40c5-9415-a3075a452543","year":2025},"citing_paper":{"arxiv_id":"2606.01161","last_updated":"2026-05-31T11:08:51Z","snapshot_observed_at":"2026-08-03T09:12:52.643122Z","submitted_at":"2026-05-31T11:08:51Z","title":"AcOrch: Accelerating Sampling-based GNN Training under CPU-NPU Heterogeneous Environments","version":1},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-06-28T16:37:20.774251Z"},"links":{"cited_paper":"/paper/2502.06982","citing_paper":"/paper/2606.01161"},"observation_digest":"sha256:101067ceed1e85f66691b7720d1a80911ddbbbd56266ccdb0fbc56a6f3072e2d","observation_id":"56ade0ac-c6ae-4299-85e1-3e64c5c9d9e1","resolution":{"observed_at":"2026-07-01T21:36:15.545052Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2502.06982","snapshot_observed_at":"2026-07-12T11:05:56.233115Z","title":"Machine learning fleet efficiency: Analyzing and optimizing large-scale Google TPU systems with ML productivity goodput, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.02558","last_updated":"2026-06-28T02:30:44Z","snapshot_observed_at":"2026-08-16T13:37:09.237441Z","submitted_at":"2026-06-28T02:30:44Z","title":"MLSYSIM: First-Principles Infrastructure Modeling for Machine Learning Systems","version":1},"reference_index":132,"source":"arxiv_source","source_observed_at":"2026-07-12T11:05:56.233115Z"},"links":{"cited_paper":"/paper/2502.06982","citing_paper":"/paper/2607.02558"},"observation_digest":"sha256:4b0970c5b86a5b124c82baafd480df659d60c891b357f86bb40c7350e194d45f","observation_id":"067f8b27-9930-4cb9-91ea-6a262abcfd68","resolution":{"observed_at":"2026-07-12T11:05:56.233115Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2502.06982","snapshot_observed_at":"2026-07-31T08:21:13.989830Z","title":"arXiv preprint arXiv:2502.06982 (2025)","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.28407","last_updated":"2026-07-30T15:55:29Z","snapshot_observed_at":"2026-08-16T14:13:42.368112Z","submitted_at":"2026-07-30T15:55:29Z","title":"A Taxonomy of Performance Metrics for the Distributed Computing Continuum","version":1},"reference_index":82,"source":"pdf_text","source_observed_at":"2026-07-31T08:21:13.989830Z"},"links":{"cited_paper":"/paper/2502.06982","citing_paper":"/paper/2607.28407"},"observation_digest":"sha256:fc90ea0fc5c3af45cc7c043459883b875545cdafc267d33e4cb239d8a888c72c","observation_id":"e937d750-074a-4c97-998a-3aa2b7833d3f","resolution":{"observed_at":"2026-07-31T08:21:13.989830Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"links":{"evidence":"/evidence","html":"/paper/2502.06982/citation-record","integrity":"/paper/2502.06982/integrity","json":"/paper/2502.06982/citation-record.json","paper":"/paper/2502.06982"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T14:14:37.095395Z","title":null,"venue":null,"work_id":"7740d90b-63cf-476c-978d-7f019de0bbc7","year":null},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.159861Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:bcb824ffc1a817fd404cb55173f7e9a8f2a2c55031e8c236cc0b75cea049427f","observation_id":"cea38c78-cb23-4e32-891e-8cfc71194c40","resolution":{"observed_at":"2026-08-08T14:14:37.099828Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:37.082698Z","title":null,"venue":null,"work_id":"02c98bd1-76b0-46dd-85b0-73a228d772e6","year":null},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.164732Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:e18aef355379bed2949d2c1e4960149b1139453bc1a96ee3d61852e075a3c343","observation_id":"0b4dea27-9028-45d8-a809-974e48dd30b7","resolution":{"observed_at":"2026-08-08T14:14:37.087597Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1605.08695","last_updated":"2016-05-31T19:46:10Z","snapshot_observed_at":"2026-08-16T23:31:40.523493Z","submitted_at":"2016-05-27T15:49:50Z","title":"TensorFlow: A system for large-scale machine learning","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1605.08695","snapshot_observed_at":"2026-08-08T14:14:35.173870Z","title":"Murray, Benoit Steiner, Paul Tucker, Vijay Vasudevan, Pete Warden, Martin Wicke, Yuan Yu, and Xiaoqiang Zheng","venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.173870Z"},"links":{"cited_paper":"/paper/1605.08695","citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:41346103e909b55b1eb05ad443bc348df4a3451f75b6b66b30b2dce069d5204a","observation_id":"ba93c145-4b2e-4506-89aa-11ff69ed5b23","resolution":{"observed_at":"2026-08-08T14:14:35.173870Z","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":"2026-08-08T14:14:37.069700Z","title":"Banning, Sumeer Bhola, Rick Buskens, Ming Chen, Xi Chen, Yoo Chung, Qin Jia, Nick Sakharov, George T","venue":null,"work_id":"721ae1eb-76ac-4e94-8056-eb78ef285ced","year":2020},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.178126Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:f28fad620f883ab907f308cb701ac2703e7e6922c73d5a87ede8f06b62f87d40","observation_id":"25394408-4efc-4f53-b8ec-bde0f8a2f269","resolution":{"observed_at":"2026-08-08T14:14:37.074250Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:37.057629Z","title":null,"venue":null,"work_id":"bfc1c364-05bf-450f-bbe1-54032d35af93","year":2024},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.182115Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:20a0beba546351789ebb279a25f38dc3e6fecdbe1652ee383795b44dd199a043","observation_id":"62504617-2bff-4208-b436-f941d37f44c7","resolution":{"observed_at":"2026-08-08T14:14:37.061577Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:35.185891Z","title":null,"venue":null,"work_id":null,"year":2009},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.185891Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:039d1d5376fe91d8e3a829ef2f45f7638c64d6dc6e15104ffc61d979148a0e61","observation_id":"31b47966-264d-4026-820a-a17fd17da18e","resolution":{"observed_at":"2026-08-08T14:14:35.185891Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2203.12533","last_updated":"2022-03-23T16:50:53Z","snapshot_observed_at":"2026-08-17T21:14:51.112405Z","submitted_at":"2022-03-23T16:50:53Z","title":"Pathways: Asynchronous Distributed Dataflow for ML","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2203.12533","snapshot_observed_at":"2026-08-08T14:14:35.190079Z","title":"Thekkath, and Yonghui Wu","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.190079Z"},"links":{"cited_paper":"/paper/2203.12533","citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:13e41ad7eb43d588f45978ed41d5e3bfb6d85829b37d912ff1e8450b2fef3e17","observation_id":"0bee877f-3334-4be6-b069-8149d26f08bf","resolution":{"observed_at":"2026-08-08T14:14:35.190079Z","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":"2026-08-08T14:14:37.042019Z","title":null,"venue":null,"work_id":"cd869a2a-cba3-4636-a94e-6b694c8f2515","year":2009},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.194477Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:7ae3b4e711b83ed66b9983bd31adc2eb58a847850aa3663c7861f0502ac58c7e","observation_id":"6aa35d2a-d3ab-4c23-80fa-86e629d65c28","resolution":{"observed_at":"2026-08-08T14:14:37.049550Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:37.028597Z","title":null,"venue":null,"work_id":"2d6b6cdc-34f3-4e5a-88c5-26fd8cedbce6","year":2008},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.198519Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:0137f1e3a11b7e8fc9f703af197136d065b29033b39d67973a5bf98bc73818e7","observation_id":"c6639a02-af13-4600-b624-732fa7140ab8","resolution":{"observed_at":"2026-08-08T14:14:37.033407Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"3095.14314","doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T14:14:36.375004Z","title":"Cooper, and Linda Torczon","venue":null,"work_id":"1c3d2e05-6246-4a52-86bd-eb87ae5454d2","year":1992},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.201998Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:14a0397e1f63f65e8952ff9b3bdbfa415be3851d6234117a8b80df11095c6eb3","observation_id":"6a921add-11f4-4482-9a0c-3307790786b9","resolution":{"observed_at":"2026-08-08T14:14:36.380989Z","resolver_source":"raw_fallback","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:37.014127Z","title":null,"venue":null,"work_id":"c514cc17-9024-4ef5-82e4-31c4c4dbc88f","year":1998},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.205703Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:a9c50b72e8a98ae16398ba0173ae12428f9a71e7a3a63dccb0e77ade926429a3","observation_id":"b2da1d18-b349-4355-abe8-ac2c17255023","resolution":{"observed_at":"2026-08-08T14:14:37.018889Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2005.14165","last_updated":"2020-07-22T19:47:17Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2020-05-28T17:29:03Z","title":"Language Models are Few-Shot Learners","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2005.14165","snapshot_observed_at":"2026-08-08T14:14:35.209953Z","title":null,"venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.209953Z"},"links":{"cited_paper":"/paper/2005.14165","citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:f4c8e5a51783cfc5c81db4960601cbaf5f92818fda3c35ee5993988ba3583295","observation_id":"3d8a7939-5e56-4b69-8c42-c3783a43f834","resolution":{"observed_at":"2026-08-08T14:14:35.209953Z","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":"2026-08-08T14:14:35.214069Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.214069Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:c481165a447ba16a002a5faf53ac3b110de2b1852ac9cf2ee8394b2d035abf14","observation_id":"06eea592-b20f-4e4b-8d92-9c1fd16bf40b","resolution":{"observed_at":"2026-08-08T14:14:35.214069Z","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":"2026-08-08T14:14:36.993445Z","title":null,"venue":null,"work_id":"9066b108-a6fc-4682-8bf2-0183598f9861","year":2024},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.221745Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:8a6988eec8ec2987ee8ccdf91fe75a513caf77420495596759e47e4a3d473448","observation_id":"9f709f81-142b-4e48-b57a-23358e43ca57","resolution":{"observed_at":"2026-08-08T14:14:36.998340Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:35.217913Z","title":null,"venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.217913Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:4ec581001b7cb47028f06e89031e06bac0ab59cb3940b0a7e6d17aa0fcf8ec40","observation_id":"3249a713-8368-4699-b59e-6da122993ffa","resolution":{"observed_at":"2026-08-08T14:14:35.217913Z","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":"2026-08-08T14:14:36.970498Z","title":null,"venue":null,"work_id":"402402c0-37e8-4491-8a41-ceeebeebabfb","year":2010},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.229279Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:03d26ef92b6bd0e0c22ff196880b4bad30fd438d0558bd7bc8feee6de9d0a2bb","observation_id":"1cdcf7a7-8a22-45ad-bcc6-363b710d7867","resolution":{"observed_at":"2026-08-08T14:14:36.973987Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:36.981692Z","title":null,"venue":null,"work_id":"58c5e2a0-fcc9-4bed-820e-e47f3a83c532","year":2012},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.225561Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:a76c7f727788ccd7917502cec85ae8e14d216b4154278050980f074b7a4bace8","observation_id":"3da094ce-d419-4776-90da-4cf7a41b2857","resolution":{"observed_at":"2026-08-08T14:14:36.986304Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:36.958983Z","title":null,"venue":null,"work_id":"15852cc8-5cc9-4f5f-8666-bb5555753ee2","year":2021},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.236342Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:ccf12d35e62568b317e3d0d6e6b92741566eaf79461826affe4030bd11db8340","observation_id":"271ae118-3e51-49a6-8c22-5fca3a4c9e90","resolution":{"observed_at":"2026-08-08T14:14:36.963328Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"3453.80819","doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T14:14:36.278700Z","title":"Emer and Douglas W","venue":null,"work_id":"281de91d-b720-469b-80d3-c3afa6908f1b","year":1984},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.232556Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:c29a32b28142622a0f7dca0ee0069093e34b266292e71a29657e7ca5fe2afa60","observation_id":"5d71858a-c111-4394-91c4-3007c6bd2bb5","resolution":{"observed_at":"2026-08-08T14:14:36.283984Z","resolver_source":"raw_fallback","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:36.936131Z","title":null,"venue":null,"work_id":"84fa5ac2-6754-4f60-b42e-624122d28652","year":2018},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.243564Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:0c371f1e007fc37f176eeed09e74fdce19276da8fc1dfc37ef3389a05c55ee0e","observation_id":"3ada0bbe-4023-4306-949b-5ad8a1d8acc4","resolution":{"observed_at":"2026-08-08T14:14:36.939914Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:36.947896Z","title":null,"venue":null,"work_id":"33136d46-c479-46c5-926a-bc75752659c9","year":2004},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.239690Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:f77867edd73da212859e96cd0fcc2c1f7a434f75c276bf669ea0284daca072cc","observation_id":"44db8799-8f3c-4dc5-badf-05346876a859","resolution":{"observed_at":"2026-08-08T14:14:36.951919Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:36.912784Z","title":null,"venue":null,"work_id":"44bf13ab-b122-400c-b7c8-4bab85ba9cb0","year":2007},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.250803Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:bfa01244014f70388f9897201961c4e7634ed80f6f55f595ef340f45d3de41ea","observation_id":"87fb889a-4061-43d3-bf8e-740bf237af1d","resolution":{"observed_at":"2026-08-08T14:14:36.916849Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:36.924720Z","title":null,"venue":null,"work_id":"7a784c83-7117-4995-87e0-897eaac1c156","year":2003},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.247043Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:0e741d43adf4f843d05ad9205c1015522c81fd039106e799288f47e77e634856","observation_id":"aeaf1fd6-7599-4aaf-8326-02135d738274","resolution":{"observed_at":"2026-08-08T14:14:36.928462Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:36.886415Z","title":null,"venue":null,"work_id":"995786ad-a087-44bf-abec-29b4cae3b20f","year":2021},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.258072Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:85bdee2050e93867d6d77d570d725021972c327abb86df91a5f66ef74fa0e6e4","observation_id":"3fb2793b-0597-426b-835a-8a363aa6a5ab","resolution":{"observed_at":"2026-08-08T14:14:36.890153Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:36.898526Z","title":"Hennessy and David A","venue":null,"work_id":"ff7d4c7a-28b4-4a4f-8b9f-9691da27fcdf","year":2019},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.254605Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:3fd240237d9df5394b9cb83dc90bfd8eff66906606c0dfe2c4af092901b81577","observation_id":"18c5a57b-994b-45ff-ae44-aeb52ba82a7c","resolution":{"observed_at":"2026-08-08T14:14:36.903414Z","resolver_source":"raw_fallback","status":"malformed_identifier"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:36.860825Z","title":null,"venue":null,"work_id":"cfe3bfd7-82ea-4270-a787-4618a54c185f","year":null},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.265555Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:4f16375f1ab26c8adf2362ef259be66f73a8086f80a661b0041a55e648946225","observation_id":"fa0088d6-7fc3-43aa-b781-393a2d03c52e","resolution":{"observed_at":"2026-08-08T14:14:36.865391Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:36.873513Z","title":null,"venue":null,"work_id":"565bd790-ed42-4901-beba-341147cfa9e5","year":2020},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.261802Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:fabef9685134d2b254abc4cd71afb0a2a55a588d1e2b188aaec9109458e91e68","observation_id":"fab075ab-53f4-40d7-8332-10a85963d716","resolution":{"observed_at":"2026-08-08T14:14:36.878769Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2304.01433","last_updated":"2023-04-20T22:25:51Z","snapshot_observed_at":"2026-08-16T15:43:01.932266Z","submitted_at":"2023-04-04T00:52:46Z","title":"TPU v4: An Optically Reconfigurable Supercomputer for Machine Learning with Hardware Support for Embeddings","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2304.01433","snapshot_observed_at":"2026-08-08T14:14:35.272822Z","title":null,"venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.272822Z"},"links":{"cited_paper":"/paper/2304.01433","citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:486824db6bef1fd8af144dd9b33c1c416da3a9bd8359b92a3ffb50121ca42a9d","observation_id":"a383b0ba-cb36-47ff-a4f3-5a3c501ef85f","resolution":{"observed_at":"2026-08-08T14:14:35.272822Z","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":"2026-08-08T14:14:35.269272Z","title":"Jouppi, Doe Hyun Yoon, Matthew Ashcraft, Mark Gottscho, Thomas B","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.269272Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:cc4ad23534d0e17188d365437a563d09c4fe67077dfc1b8c8e600625c9987c6e","observation_id":"a04529fe-3d6e-4562-a63d-1f61baacdcc2","resolution":{"observed_at":"2026-08-08T14:14:35.269272Z","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":"2008.45362","doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T14:14:36.109190Z","title":null,"venue":null,"work_id":"2f50beec-bb15-40cf-8a79-e99a44cb033e","year":2008},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.284853Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:679766b91f8229ee324787b5a095ae2fc84ec1d2bc111d219463abd3e81104ea","observation_id":"f9abe851-84b4-4daf-96cf-3ffaf96d86d2","resolution":{"observed_at":"2026-08-08T14:14:36.116560Z","resolver_source":"raw_fallback","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:36.846332Z","title":"Jouppi, Cliff Young, Nishant Patil, David A","venue":null,"work_id":"1ab8a017-5dae-410d-991c-f8ff5f0d9803","year":null},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.276616Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:974785f096b32a67725c039c2184273985d3ab70601fc03d4f897aa3ee86f475","observation_id":"7e35396e-d937-4097-995a-bb982854c260","resolution":{"observed_at":"2026-08-08T14:14:36.852533Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:36.821777Z","title":null,"venue":null,"work_id":"e257bb8d-32a9-4022-9037-562e7b8fa730","year":2012},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.291882Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:0eb65317a98fc63948cd443b0572f42682d9bc53557730fec7f0bce5a45af864","observation_id":"04723395-cbe5-4fd1-9ddc-be9bcb5f78dc","resolution":{"observed_at":"2026-08-08T14:14:36.825662Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:36.806831Z","title":null,"venue":null,"work_id":"61dc47b4-9030-4746-97af-97da11024177","year":2022},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.295663Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:0f26f86844406392686f95328f9a1d422f8b56a0ac69e2c85c70978f332627b8","observation_id":"40174fd1-2797-43c3-ae8e-0389fe5574ee","resolution":{"observed_at":"2026-08-08T14:14:36.811728Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:36.834340Z","title":null,"venue":null,"work_id":"dfaea593-9822-4439-9326-be23734f069c","year":2015},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.288453Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:67ff8efe2636891cc68757fa1d3805d483fdbf3a77a0253fe842df3024e094c8","observation_id":"8aa90ba5-3253-4e87-9805-7f31fa76a548","resolution":{"observed_at":"2026-08-08T14:14:36.838478Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:36.782211Z","title":null,"venue":null,"work_id":"a345a1ff-8187-4939-b1a9-0e7635f93e37","year":null},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.303755Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:d2786935da391c9f514e8196dbfd88bf656f207046501555add25dd2c901de2f","observation_id":"4ac63b1a-3cd3-4886-8bbc-eb203aa1bbfd","resolution":{"observed_at":"2026-08-08T14:14:36.786613Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:36.758712Z","title":null,"venue":null,"work_id":"c0267f73-4236-487e-8d10-c45268d6588f","year":2014},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.312034Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:c258367aea6656eaa885379dacf0b7c350fc38d126fee0942c99b533b7e44a22","observation_id":"68f558cb-ed27-41ae-bfcc-4886d4f215bd","resolution":{"observed_at":"2026-08-08T14:14:36.762357Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:36.794813Z","title":null,"venue":null,"work_id":"774bacac-0a34-4862-acfd-91c8c7aefe62","year":2021},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.300028Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:9da45a18d265b885575b60a100e54ed1b776a81adc26be54bbfd872cfae3c740","observation_id":"edde4ab9-8bd1-46dd-8c57-a6843fd05f34","resolution":{"observed_at":"2026-08-08T14:14:36.799045Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:36.724020Z","title":null,"venue":null,"work_id":"cb6fee4a-520a-4539-9779-6916322d1a09","year":2020},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.323018Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:1497e08e388dfd2c07f3188bfb281ceeb21604fd1df004c4a9123447cdc673e2","observation_id":"cd04e0e4-912f-47fb-a144-b30e5033e5a5","resolution":{"observed_at":"2026-08-08T14:14:36.728319Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:35.326881Z","title":"Mustafa Rafique, Franck Cappello, and Bogdan Nicolae","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.326881Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:d09af11d3c0d66420a5a8d72cdcdd5f71032226a5212dbd7ce00c3b7d3356d62","observation_id":"ab82423b-1998-4c8e-bc28-340e0a3c3a3e","resolution":{"observed_at":"2026-08-08T14:14:35.326881Z","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":"2026-08-08T14:14:36.711530Z","title":null,"venue":null,"work_id":"241e2ed1-ff55-41e1-bc74-49617fa1563c","year":2005},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.330529Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:8c00f3ae45b49e53a5a93dc21ca96745138d25805a25f5595b9a44e78d144819","observation_id":"b89ee389-a29c-4bf0-adc7-c68a0b78d1f3","resolution":{"observed_at":"2026-08-08T14:14:36.716124Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:35.315520Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.315520Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:00fd8a23fd9b8a3a2d9527fff64e9dc6630623d57ded2e1794aa6c8d77d46f6a","observation_id":"1fd3196a-60c5-4b1d-b348-6451670d219c","resolution":{"observed_at":"2026-08-08T14:14:35.315520Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1906.00091","last_updated":"2019-05-31T21:51:16Z","snapshot_observed_at":"2026-08-14T16:22:46.075072Z","submitted_at":"2019-05-31T21:51:16Z","title":"Deep Learning Recommendation Model for Personalization and Recommendation Systems","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1906.00091","snapshot_observed_at":"2026-08-08T14:14:35.337895Z","title":null,"venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.337895Z"},"links":{"cited_paper":"/paper/1906.00091","citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:3a2f5d06d28cb039b6bafbd16e976a3eadbaff684965a08afd4cce1f2330ca24","observation_id":"d8522ad1-6365-45dc-9a64-2d62889e1620","resolution":{"observed_at":"2026-08-08T14:14:35.337895Z","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":"2026-08-08T14:14:35.341659Z","title":"Wozniak, George Bosilca, Matthieu Dorier, and Franck Cappello","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.341659Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:e75829fef66f63f9a9bedb7181a0c8e6ac72f38b34c455ae6db51edc71a19768","observation_id":"46a99af2-e73d-43ec-bd45-7fc432253e3d","resolution":{"observed_at":"2026-08-08T14:14:35.341659Z","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":"2026-08-08T14:14:36.698837Z","title":"Li, Ryan McElroy, Mike Paleczny, Daniel Peek, Paul Saab, David Stafford, Tony Tung, and Venkateshwaran Venkataramani","venue":null,"work_id":"7d103ab0-2080-4a1b-b5ba-9e8664132a82","year":2013},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":44,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.345317Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:4b323037d57768994797dd934cbf4071e1cb19ab45152901d343287deed8b313","observation_id":"d7582782-e7b2-40e7-aa22-f70f980fc257","resolution":{"observed_at":"2026-08-08T14:14:36.703233Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"2020.00075","doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T14:14:35.856689Z","title":null,"venue":null,"work_id":"1be2fb53-949d-4245-8a8e-6d4970ab2be4","year":2020},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":45,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.348829Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:30d53e021b57a41b83e58c8d55b7d07ad0de7e2c250a34fd98917bce071b27aa","observation_id":"e0bd02fb-8097-4562-94ec-647cb41c74ff","resolution":{"observed_at":"2026-08-08T14:14:35.862508Z","resolver_source":"raw_fallback","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2101.12127","last_updated":"2021-02-23T22:56:12Z","snapshot_observed_at":"2026-08-16T18:49:18.604576Z","submitted_at":"2021-01-28T17:16:46Z","title":"tf.data: A Machine Learning Data Processing Framework","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2101.12127","snapshot_observed_at":"2026-08-08T14:14:35.334066Z","title":"Murray, Jiri Simsa, Ana Klimovic, and Ihor Indyk","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":46,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.334066Z"},"links":{"cited_paper":"/paper/2101.12127","citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:e6fad9c6ec47f3c15355190cd955b38e61016320f5239803d817d0f1f2e36ab2","observation_id":"0ecf53de-5d5d-4554-aa1e-9b436101f1bf","resolution":{"observed_at":"2026-08-08T14:14:35.334066Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1912.01703","last_updated":"2019-12-03T22:06:05Z","snapshot_observed_at":"2026-07-06T08:41:49.632205Z","submitted_at":"2019-12-03T22:06:05Z","title":"PyTorch: An Imperative Style, High-Performance Deep Learning Library","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1912.01703","snapshot_observed_at":"2026-08-08T14:14:35.356616Z","title":null,"venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":47,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.356616Z"},"links":{"cited_paper":"/paper/1912.01703","citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:a093694074abbce018ff20f85992957dc9f986a78be1baad2fe2196457005c0b","observation_id":"4364c9d7-070c-4f5b-9c59-84e7e8ba1517","resolution":{"observed_at":"2026-08-08T14:14:35.356616Z","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":"2026-08-08T14:14:36.674851Z","title":null,"venue":null,"work_id":"83949f02-0477-476c-bd0e-504a63bd42c7","year":2021},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":48,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.360582Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:c12eac182bc13592db764e35731d046436cca1bc63d29a49b20acb7abbf0f81f","observation_id":"f925e958-84f0-4b51-ac10-d7dce73ce6dd","resolution":{"observed_at":"2026-08-08T14:14:36.678965Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:35.364656Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":49,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.364656Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:e54c808ee488da0f8389fbb081d5116c9f4fee3317f752cbd40392ba2229b422","observation_id":"47937785-8b0a-4ee3-ab3f-aeec2bed55c5","resolution":{"observed_at":"2026-08-08T14:14:35.364656Z","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":"2024.34094","doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T14:14:35.757656Z","title":null,"venue":null,"work_id":"fee981ba-b386-4754-9fa7-778e8d50a624","year":2024},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":50,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.373018Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:c23f1d3414a3518b0b1de2033c66b0309fdf86596ba8854844ed2009a843d461","observation_id":"f0baefb9-e41d-4293-a820-61eca9060050","resolution":{"observed_at":"2026-08-08T14:14:35.763370Z","resolver_source":"raw_fallback","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:36.687347Z","title":null,"venue":null,"work_id":"cbc44844-d1bb-40da-bb1f-4274dd7a72a1","year":null},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":51,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.352212Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:f536bb43789cf0111b0978899a7685a0d1701a43a5e26ad6fc4871117793637d","observation_id":"c3a6bd37-7003-4a44-a5c7-ce46522edf43","resolution":{"observed_at":"2026-08-08T14:14:36.691185Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:36.630898Z","title":null,"venue":null,"work_id":"7c1d3355-af69-40d8-9993-93836af6293b","year":null},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":52,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.380584Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:08808c0f4ba88be4615c43a261103c902ed432eee24d230f147d65502a78b6a3","observation_id":"fe464644-1d9c-4018-b17d-0f4c6115eb9f","resolution":{"observed_at":"2026-08-08T14:14:36.634749Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:36.618265Z","title":null,"venue":null,"work_id":"c132df6a-6d1d-4119-a510-29b6d44a4bca","year":2009},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":53,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.383699Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:c781b7b68eb84c0f4cbdb08b683da2ddc281108f842f6a14bac49ed0c30e5db9","observation_id":"890106eb-518b-4e56-ba70-5d9027342048","resolution":{"observed_at":"2026-08-08T14:14:36.622281Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1701.06538","last_updated":"2017-01-23T18:10:00Z","snapshot_observed_at":"2026-08-13T11:35:07.866136Z","submitted_at":"2017-01-23T18:10:00Z","title":"Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1701.06538","snapshot_observed_at":"2026-08-08T14:14:35.387292Z","title":null,"venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":54,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.387292Z"},"links":{"cited_paper":"/paper/1701.06538","citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:78d913386502ac3273e48de515e9c548272a09c431aa8078c7c7f0cb98f8b6e1","observation_id":"d26cc555-a91d-4763-94ad-c2d91c7904f3","resolution":{"observed_at":"2026-08-08T14:14:35.387292Z","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":"2026-08-08T14:14:36.654616Z","title":"In Proceedings of Ma- chine Learning and Systems , D","venue":null,"work_id":"60f50d26-d10f-4586-bc34-e959f0f6c8fb","year":2023},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":55,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.368606Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:e3843364664b3599aa8f726c7ca0ae61202cef92a812f386c121dd4b2e5aff1c","observation_id":"7a8fa365-89e1-4dae-a7c3-a0c6220d91bb","resolution":{"observed_at":"2026-08-08T14:14:36.659203Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:36.593123Z","title":null,"venue":null,"work_id":"38656d99-dfe5-4e1c-8537-4a22eab6384d","year":2019},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":56,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.399359Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:b0470b0888f8094ed64faa49ded7d97e46650ee00f744913cfa9a4bda9e8127e","observation_id":"5906e66f-f389-487c-b4ce-db3b8e78b1b2","resolution":{"observed_at":"2026-08-08T14:14:36.596928Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:36.642444Z","title":null,"venue":null,"work_id":"5887ea51-58cd-4f19-9a61-1d6fcedd265d","year":2015},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":57,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.376716Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:2af6d7b021012702cf4604aae21c51923283db575b299f7910f1ecd623e6f815","observation_id":"ec8ef137-7b5d-4351-ace9-ab1db94c451f","resolution":{"observed_at":"2026-08-08T14:14:36.646435Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2312.11805","last_updated":"2025-05-09T21:04:06Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2023-12-19T02:39:27Z","title":"Gemini: A Family of Highly Capable Multimodal Models","version":5},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2312.11805","snapshot_observed_at":"2026-08-08T14:14:35.406525Z","title":null,"venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":58,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.406525Z"},"links":{"cited_paper":"/paper/2312.11805","citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:bdfde1bb9a89c4d67365d6ab4958acc9d9d63a2022aebb6b53629ac1286641ea","observation_id":"08807a89-72ff-4ac9-bd5d-27972fda8f58","resolution":{"observed_at":"2026-08-08T14:14:35.406525Z","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":"2026-08-08T14:14:36.568314Z","title":null,"venue":null,"work_id":"24ca1aa4-2da5-4e14-8a53-e5acab0b756b","year":2023},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":59,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.410326Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:f429d0e5c360f9743a25dde2d0bcfdf8988e476785fb284046e7db2074ac9bd7","observation_id":"853f7608-fb94-49a3-bd3f-3d51f898e901","resolution":{"observed_at":"2026-08-08T14:14:36.572563Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:36.555230Z","title":null,"venue":null,"work_id":"26c982ed-a8cc-4449-8cf7-80a5a59c6a80","year":2008},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":60,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.413646Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:63c9823728baec806ae44ec4f0f146d0b87158c8512ac4b3bfcb541bb94f7f2c","observation_id":"904ed2ba-818e-456a-b955-e57284ea0b3a","resolution":{"observed_at":"2026-08-08T14:14:36.559271Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:36.604515Z","title":null,"venue":null,"work_id":"40a47d3e-0dde-420e-bed8-c7335f72be2e","year":null},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":61,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.391018Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:d1c835d038254b8907c52ab399c84c3dd8e8ccbf85465535136c26eac26dafdf","observation_id":"ac4921a7-110b-4bfb-85dc-a81c9c424164","resolution":{"observed_at":"2026-08-08T14:14:36.609749Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:36.535685Z","title":"Korupolu, David Oppenheimer, Eric Tune, and John Wilkes","venue":null,"work_id":"56170e9a-0bd6-4409-8a2b-01b43012640d","year":2015},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":62,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.422067Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:47728b2f2b5d099afa46f58168e3725733c6b1f9a23be53d0d61cc2f60894d52","observation_id":"1c7deb3f-fc0a-47f7-aeee-3d452d9d5be2","resolution":{"observed_at":"2026-08-08T14:14:36.539676Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:36.523667Z","title":null,"venue":null,"work_id":"3f7a3c8f-f3bf-4aa1-bcf5-aa9726bea2bb","year":2022},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":63,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.425471Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:2ea4e4ad36ea15f5df6016b690a33273921431d234541a9522f44d5311d92ff3","observation_id":"1bdd4758-21ba-4026-91b0-1158d0faa395","resolution":{"observed_at":"2026-08-08T14:14:36.527478Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:36.581023Z","title":null,"venue":null,"work_id":"e09943f9-394f-4c54-9027-333c8566925c","year":2016},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":64,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.402764Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:6fed338c02038307791adffefcb473f67c11b82838981af6bd93ec1afe64bc0d","observation_id":"6118bd45-25b0-4fc1-914f-1ddd3305fad0","resolution":{"observed_at":"2026-08-08T14:14:36.585251Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2102.10423","last_updated":"2022-10-11T16:02:43Z","snapshot_observed_at":"2026-08-16T18:44:12.549860Z","submitted_at":"2021-02-20T19:25:09Z","title":"An Evaluation of Edge TPU Accelerators for Convolutional Neural Networks","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2102.10423","snapshot_observed_at":"2026-08-08T14:14:35.432953Z","title":null,"venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":65,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.432953Z"},"links":{"cited_paper":"/paper/2102.10423","citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:75bd33fd9da02d2c3fb34c5b0b3cab9d300df811db5e8a30bd80611146662573","observation_id":"fdf12497-143c-417d-9602-cacddbb99a31","resolution":{"observed_at":"2026-08-08T14:14:35.432953Z","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":"2026-08-08T14:14:36.510899Z","title":"Yoo, Morris A","venue":null,"work_id":"3b49d28e-b1e7-4e91-995d-a11b4cf22ce6","year":2003},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":66,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.436777Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:f97bbb4f82d94430ccc75e8a77a550c21b1b39144d851b6f7ac05446af0a2717","observation_id":"39d43346-6e2b-489e-b7cb-18c1aa711eda","resolution":{"observed_at":"2026-08-08T14:14:36.515371Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2108.09373","last_updated":"2022-04-22T23:51:04Z","snapshot_observed_at":"2026-08-16T18:01:57.154854Z","submitted_at":"2021-08-20T21:09:34Z","title":"Understanding Data Storage and Ingestion for Large-Scale Deep Recommendation Model Training","version":4},"cited_work":{"arxiv_id":"2108.09373","doi":null,"metadata_source":"pith","pith_arxiv_id":"2108.09373","snapshot_observed_at":"2026-08-08T14:14:35.562320Z","title":"Understanding Data Storage and Ingestion for Large-Scale Deep Recommendation Model Training","venue":"cs.DC","work_id":"5a9c0b67-0b85-41b3-873f-80c07c0eb9a7","year":2021},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":67,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.440366Z"},"links":{"cited_paper":"/paper/2108.09373","citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:ba173040e367c193fadec5d45e190186d09e567352b133ed02409a17bae25adc","observation_id":"10ca42e5-1ffb-4ccd-ba5b-00e93d4f239b","resolution":{"observed_at":"2026-08-08T14:14:35.568865Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:35.417476Z","title":null,"venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":68,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.417476Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:70465ee241224e37ad86bf1d194cc530f231086627cf84a75f7c9cd4f25acd78","observation_id":"3c795c23-a7ed-4002-a3c5-1ffc392fa3d6","resolution":{"observed_at":"2026-08-08T14:14:35.417476Z","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":"2026-08-08T14:14:36.498571Z","title":null,"venue":null,"work_id":"d4f32faa-ac89-495a-b024-875add003cf4","year":2024},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":69,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.448676Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:78fbef5655efecce52ca40be41a99907196c7d5e2918fa4098be2768e506760a","observation_id":"db9cc7a8-a2bb-4a3e-ab44-a1e8936db368","resolution":{"observed_at":"2026-08-08T14:14:36.502598Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-08T14:14:35.429521Z","title":null,"venue":null,"work_id":null,"year":2009},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":71,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.429521Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:714d20c94d72a90600cb6dac5283a238b488ec9d25e07e28cf2f1251c8591a53","observation_id":"758981c4-9a02-457b-af2b-5a42ce20c145","resolution":{"observed_at":"2026-08-08T14:14:35.429521Z","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":"2026-08-08T14:14:35.444334Z","title":null,"venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":75,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.444334Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:2b4df423aa13bc827da48f2a706952ced38adf3756d653b10a3d479c074cdcd2","observation_id":"fc3ae17a-213b-4917-98b6-7e871893365c","resolution":{"observed_at":"2026-08-08T14:14:35.444334Z","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":"2026-08-08T14:14:36.737376Z","title":"In Proceedings of the 44th annual IEEE/ACM International Symposium on Microarchitecture","venue":null,"work_id":"30109c51-3dd0-42d1-80f2-3f7576f489f9","year":null},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":2011,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.319225Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:e1d441784456f7ae35807b53822f0301072bada7b0dc00122c9ee980d6f87472","observation_id":"2dfe1f75-e9ed-41f0-a3d4-35e47168f5fb","resolution":{"observed_at":"2026-08-08T14:14:36.741531Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1603.04467","last_updated":"2016-03-16T16:57:12Z","snapshot_observed_at":"2026-08-14T22:06:04.800050Z","submitted_at":"2016-03-14T20:50:20Z","title":"TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1603.04467","snapshot_observed_at":"2026-08-08T14:14:35.169430Z","title":"arXiv:1603.04467 [cs.DC] https://arxiv.org/abs/1603.04467","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":2016,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.169430Z"},"links":{"cited_paper":"/paper/1603.04467","citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:80ff7db8e42bd48f9ff6ab2bb651a89c5434b3e060c00ae27941eb9a413a5821","observation_id":"a8f6a0e4-5c66-44ee-ab88-74857485dbf5","resolution":{"observed_at":"2026-08-08T14:14:35.169430Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1704.04760","last_updated":"2017-04-16T12:07:54Z","snapshot_observed_at":"2026-08-14T21:06:33.189868Z","submitted_at":"2017-04-16T12:07:54Z","title":"In-Datacenter Performance Analysis of a Tensor Processing Unit","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1704.04760","snapshot_observed_at":"2026-08-08T14:14:35.280728Z","title":"CoRR abs/1704.04760 (2017)","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":2017,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.280728Z"},"links":{"cited_paper":"/paper/1704.04760","citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:32c0fce6102916c01f9b086b6ea0116f7e9051fbc6ce90def87ab503291f0864","observation_id":"82e0ff50-14c6-4299-ae00-bc1275b92bbc","resolution":{"observed_at":"2026-08-08T14:14:35.280728Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2010.02075","last_updated":"2020-10-05T15:12:52Z","snapshot_observed_at":"2026-08-16T19:15:48.050191Z","submitted_at":"2020-10-05T15:12:52Z","title":"Learned Hardware/Software Co-Design of Neural Accelerators","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2010.02075","snapshot_observed_at":"2026-08-08T14:14:35.395499Z","title":"CoRR abs/2010.02075 (2020)","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":2020,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.395499Z"},"links":{"cited_paper":"/paper/2010.02075","citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:e2585178121ce6f44a78457632a6b41801d9ae53cb95dde17bc78953be3ee7c7","observation_id":"39ad9462-27af-42f6-a0ae-b7ac694bcc37","resolution":{"observed_at":"2026-08-08T14:14:35.395499Z","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":"2026-08-08T14:14:36.770136Z","title":"In International Conference on High Performance Computing","venue":null,"work_id":"b173ec8b-c141-4816-8b4a-fa293c3153d9","year":null},"citing_paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput","version":2},"reference_index":2023,"source":"pdf_text","source_observed_at":"2026-08-08T14:14:35.308185Z"},"links":{"citing_paper":"/paper/2502.06982"},"observation_digest":"sha256:affd2fa43d014073dba8abf69399dba65c9e03cf9026a253dc0216919a4bf745","observation_id":"2ba9a89e-3730-4418-9db9-519699aa3567","resolution":{"observed_at":"2026-08-08T14:14:36.774606Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+00:00","source":"retraction_watch"}],"state":"measured"}}],"paper":{"arxiv_id":"2502.06982","last_updated":"2025-04-20T08:02:36Z","latest_version":2,"primary_category":"cs.LG","snapshot_observed_at":"2026-08-17T20:29:38.596319Z","submitted_at":"2025-02-10T19:20:02Z","title":"Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput"},"reference_resolution":{"displayed":76,"state_counts":{"malformed_identifier":1,"metadata_mismatch":4,"parse_uncertain":0,"unresolved":61,"verified_exact":2,"verified_fuzzy":8},"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":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+00:00","source":"retraction_watch"}],"thesis":"As of 19 August 2026, this Paper Citation Record lists 76 of 76 outbound references and 4 inbound Pith citation observations for arXiv:2502.06982."}