{"as_of":"2026-08-15T19:20:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:652dc8a3711afee5404d65c07a1c8c8540563287c5e81ca42361684add017fc9","coverage":[{"denominator":40,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":40,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-06T23:26:57.657461Z","state":"measured"},{"denominator":40,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":40,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-15T06:32:42.880941+00:00","state":"measured"},{"denominator":0,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":0,"source":"paper_references, paper_reference_links","source_observed_at":null,"state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"cited_works","source_observed_at":null,"state":"measured"}],"external_citation_measurements":[],"inbound":[],"links":{"evidence":"/evidence","html":"/paper/2506.18193/citation-record","integrity":"/paper/2506.18193/integrity","json":"/paper/2506.18193/citation-record.json","paper":"/paper/2506.18193"},"outbound":[{"citation":{"cited_paper":{"arxiv_id":"2105.04906","last_updated":"2022-01-28T12:23:37Z","snapshot_observed_at":"2026-08-12T14:02:29.797842Z","submitted_at":"2021-05-11T09:53:21Z","title":"VICReg: Variance-Invariance-Covariance Regularization for Self-Supervised Learning","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2105.04906","snapshot_observed_at":"2026-08-06T23:26:57.450855Z","title":"Vicreg: Variance-invariance-covariance regularization for self-supervised learning","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":1,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.450855Z"},"links":{"cited_paper":"/paper/2105.04906","citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:9210564c6552ada60ee3c89e22a38c62d9c2232023a719bb13dc5e250e57b30d","observation_id":"c388f0b2-1b06-4380-8cf2-6d23c7125f04","resolution":{"observed_at":"2026-08-06T23:26:57.450855Z","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":"2026-08-06T23:26:58.316845Z","title":"A simple framework for contrastive learning of visual representations","venue":null,"work_id":"1c82dac1-19a9-448e-bc52-347f0088bc13","year":2020},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":2,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.456545Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:f48c094ab0714e910d027885f5e61ecdd225c4ab8d585c7c49d609912672a0c9","observation_id":"be30775a-80c1-4fa4-a002-b92f26803bb2","resolution":{"observed_at":"2026-08-06T23:26:58.322084Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T23:26:57.462358Z","title":"Exploring simple siamese representation learning","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":3,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.462358Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:f949450628aae983c6d6a9d74aef30260e8ede1aa0c65829fc6917d5e9e0235e","observation_id":"7a0c6306-74e8-4d03-bce5-83dbc06ea3c1","resolution":{"observed_at":"2026-08-06T23:26:57.462358Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2003.04297","last_updated":"2020-03-09T17:56:49Z","snapshot_observed_at":"2026-07-06T09:03:25.467987Z","submitted_at":"2020-03-09T17:56:49Z","title":"Improved Baselines with Momentum Contrastive Learning","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2003.04297","snapshot_observed_at":"2026-08-06T23:26:57.467697Z","title":"Improved baselines with momentum contrastive learning","venue":null,"work_id":null,"year":2003},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":4,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.467697Z"},"links":{"cited_paper":"/paper/2003.04297","citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:cec3abb9531d3241890b40f2ef73354e75016439f57e09d26bf40394cf9a2572","observation_id":"784545d4-a8b4-4037-8080-9da872fa051b","resolution":{"observed_at":"2026-08-06T23:26:57.467697Z","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":"2026-08-06T23:26:57.472885Z","title":"Cover and Joy A","venue":null,"work_id":null,"year":2006},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":5,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.472885Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:7fbe9be70e6a11a5467eb3b55ad560a6b58bfc195dbb56478f259509e57449ca","observation_id":"32e81805-7175-4a6d-b78e-1a813abbfcc7","resolution":{"observed_at":"2026-08-06T23:26:57.472885Z","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":"2026-08-06T23:26:57.477781Z","title":"Deep residual learning for image recognition","venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":6,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.477781Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:56c5c78a28acc5a279e6d4cf3763f9e7cba8a1ae344bb0bfd3927dd0e2c63a60","observation_id":"08b738f5-a120-4607-9dc8-5670e287677a","resolution":{"observed_at":"2026-08-06T23:26:57.477781Z","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":"2026-08-06T23:26:57.488270Z","title":"Momentum contrast for unsupervised visual representation learning","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":7,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.488270Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:2da06ba28f1f5ad5eb5c65d5a05a2d58e23500d50bad9d37745df4fb5501d442","observation_id":"d0a78dba-7271-4fda-bf92-07a8f3786f66","resolution":{"observed_at":"2026-08-06T23:26:57.488270Z","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":"2026-08-06T23:26:58.258357Z","title":"Realizing synchronized parameter updating, dynamic layer accumulation, and forward shortcuts in supervised contrastive parallel learning","venue":null,"work_id":"be57233d-b814-4bbe-bffb-87a6fa742ca5","year":2022},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":8,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.493838Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:eab860a6295fdcd38495b39c417558a5fabd7b8f25644ca7811e524da0a2ff55","observation_id":"e82299ff-2db8-4652-b5f2-fdeed3903fee","resolution":{"observed_at":"2026-08-06T23:26:58.263815Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T23:26:58.240849Z","title":"The vanishing gradient problem during learning recurrent neural nets and problem solutions","venue":null,"work_id":"b331e272-f6c1-4b6b-bb96-d5d09757c527","year":1998},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":9,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.498485Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:a72dda2ae5d92970eaeb361fb549cdc71ca31848d8c394705161fd64642a2ca3","observation_id":"4dcd5582-e147-4f6c-b586-29347b88fb07","resolution":{"observed_at":"2026-08-06T23:26:58.247747Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T23:26:57.503717Z","title":"Gpipe: Efficient training of giant neural networks using pipeline parallelism","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":10,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.503717Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:42d5c4ff8980a5200d992bbcecadeabe4826f08b8ff5d767052b5a36e1429d59","observation_id":"fdc4c9c2-c22a-452e-906b-5473b8124927","resolution":{"observed_at":"2026-08-06T23:26:57.503717Z","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":"2026-08-06T23:26:57.509106Z","title":"Batch normalization: Accelerating deep network training by reducing internal covariate shift","venue":null,"work_id":null,"year":2015},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":11,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.509106Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:5258a5d38318d80618bfd105293ab3cd2cc7ee3273f462da8e168a423dfe823a","observation_id":"2728d8a9-bb81-4295-b989-4564a9edbec6","resolution":{"observed_at":"2026-08-06T23:26:57.509106Z","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":"2026-08-06T23:26:58.200565Z","title":"Decoupled neural interfaces using synthetic gradients","venue":null,"work_id":"27e29bdf-a2fe-4101-a813-b307d5311b07","year":2017},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":12,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.514659Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:063a688f1e042fbb09693db37d5010e0a674d35e8c63441e70e42980d8bdad37","observation_id":"3a6da181-4b7f-457d-a2c8-41d2e86e94b9","resolution":{"observed_at":"2026-08-06T23:26:58.205482Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T23:26:58.180898Z","title":"Scaling up visual and vision-language representation learning with noisy text supervision","venue":null,"work_id":"950e2fd9-993e-4c92-8401-1776d20dc831","year":2021},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":13,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.519927Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:c3a4a367757d2f4e5f9b54f807efe163dec4e6edcf560f457e7bb01c2b105b73","observation_id":"242ffc12-a2b4-46ce-a093-c1d5d4d59350","resolution":{"observed_at":"2026-08-06T23:26:58.189312Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T23:26:58.161524Z","title":"Beyond data and model parallelism for deep neural networks","venue":null,"work_id":"7d59d525-37fc-4f81-8693-753f434cafcc","year":2019},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":14,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.524598Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:acfa9686756a11e862b953b7e4a0a2bb53c24340648829686e7645c8b4423e9e","observation_id":"fe51f5de-4382-40eb-bb94-b50c88b6c61c","resolution":{"observed_at":"2026-08-06T23:26:58.167616Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2110.09348","last_updated":"2022-04-23T16:44:20Z","snapshot_observed_at":"2026-08-15T03:32:28.162275Z","submitted_at":"2021-10-18T14:22:19Z","title":"Understanding Dimensional Collapse in Contrastive Self-supervised Learning","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.09348","snapshot_observed_at":"2026-08-06T23:26:57.529622Z","title":"Understanding dimensional collapse in contrastive self-supervised learning","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":15,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.529622Z"},"links":{"cited_paper":"/paper/2110.09348","citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:8440b26866bb4fb82451f7cb08d7b7f5c8d108f527c26a52966dca4eae3cdf7d","observation_id":"49fab7b8-4434-4fb6-841b-ee94661c9887","resolution":{"observed_at":"2026-08-06T23:26:57.529622Z","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":"2026-08-06T23:26:58.143884Z","title":"Associated learning: Decomposing end-to-end backpropagation based on autoencoders and target propagation","venue":null,"work_id":"52a59fd0-8111-4f84-9eb6-6255647983c4","year":2021},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":16,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.535725Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:0e1b5147662245fffceac6a2c61e899147cc7ab5616be92a9bb11ff04a920d85","observation_id":"2ce60da5-11a2-4fbc-abcf-dfadacb3f077","resolution":{"observed_at":"2026-08-06T23:26:58.149525Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T23:26:57.541027Z","title":"Supervised contrastive learning","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":17,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.541027Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:8d5561b3db46127415124954200865b1235f587fbcfc7514aadeba631df93a8f","observation_id":"8bc937a6-9fcd-486e-8d5f-1a950da3c547","resolution":{"observed_at":"2026-08-06T23:26:57.541027Z","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":"2026-08-06T23:26:58.111115Z","title":"Deeply-supervised nets","venue":null,"work_id":"789df0bb-c785-41b0-b5a0-5d450d7234fa","year":2015},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":18,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.546520Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:b4ba552ed59adc345743955f973ef17174f65352bbddb7a782c777050609e76b","observation_id":"153bf6f5-f95e-4bf9-a030-c85e0ef45a5d","resolution":{"observed_at":"2026-08-06T23:26:58.118048Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T23:26:58.093628Z","title":"Self-organization in a perceptual network","venue":null,"work_id":"134bd78b-6db1-4fd7-934c-4f64a2c674a9","year":1988},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":19,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.551019Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:f6dbbeb4b9b9c2584e76c5a19ab38f202b4052c10fc6fbcaedd5bf0b856be6a1","observation_id":"3af4fdac-b2c4-45fc-91d6-3dcdb32897d8","resolution":{"observed_at":"2026-08-06T23:26:58.099196Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T23:26:58.077217Z","title":"Pipedream: Generalized pipeline parallelism for dnn training","venue":null,"work_id":"f23ea55d-4377-4e0f-acbc-e28b2311ce82","year":2019},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":20,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.556575Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:80d9d21d9acbf42b06ed9b6a4eae50f5280a29ad4b7d2e440add44d704e22dee","observation_id":"dee2e877-933f-4b7c-ba2a-112565c41671","resolution":{"observed_at":"2026-08-06T23:26:58.082012Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T23:26:58.062210Z","title":"Training neural networks with local error signals","venue":null,"work_id":"b00d4535-2625-4d00-bc16-54b47fe63c0d","year":2019},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":21,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.561148Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:21cc551b588bff857d4017dac4c78faa6026fd96b0ded9e0f8b954ba9b93bdbc","observation_id":"012f512a-8fc0-4a50-be4e-f0b8185696eb","resolution":{"observed_at":"2026-08-06T23:26:58.066682Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1807.03748","last_updated":"2019-01-22T18:47:12Z","snapshot_observed_at":"2026-08-14T18:53:38.574749Z","submitted_at":"2018-07-10T16:52:11Z","title":"Representation Learning with Contrastive Predictive Coding","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1807.03748","snapshot_observed_at":"2026-08-06T23:26:57.565665Z","title":"Representation learning with contrastive predictive coding","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":22,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.565665Z"},"links":{"cited_paper":"/paper/1807.03748","citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:0de22ad60c9233b95b49163f815fdf89aeba84eb3591d069881eea49f0ed11ca","observation_id":"8ce6b912-27a8-43bb-8b4a-360fe840e197","resolution":{"observed_at":"2026-08-06T23:26:57.565665Z","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":"2026-08-06T23:26:58.046331Z","title":"Self-supervised learning with an information maximization criterion","venue":null,"work_id":"de1c2b37-e623-4227-9dbc-f8672a0035e3","year":2022},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":23,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.570723Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:f34ab0f22a7072501a9fb462884252ec937524d6794116b546f3d0430e9dfa1b","observation_id":"22345f68-9e7b-4d74-86bd-dfc5ae86257f","resolution":{"observed_at":"2026-08-06T23:26:58.051244Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T23:26:58.028711Z","title":"Glove: Global vectors for word representation","venue":null,"work_id":"1cf32b1e-460e-4ffc-9817-daa4229c991e","year":2014},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":24,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.575506Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:2fcfeb1db991a25309ebaaee01db571e6092fe9a9aeb175f6ebb30c4627daf96","observation_id":"df07c045-0e8e-455b-9968-7d1d510601a8","resolution":{"observed_at":"2026-08-06T23:26:58.034829Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T23:26:57.579992Z","title":"Learning transferable visual models from natural language supervision","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":25,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.579992Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:24b07260338b540eac3d418e9f729f431f0f71e32c5149df64891de4f3957bdf","observation_id":"421aacf7-3a6b-47f2-bafb-b38ba406ec97","resolution":{"observed_at":"2026-08-06T23:26:57.579992Z","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":"2026-08-06T23:26:57.999429Z","title":"Measuring the effects of data parallelism on neural network training","venue":null,"work_id":"cca4ffd5-c9bc-453a-bf53-d80944b9a2d3","year":2019},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":26,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.584931Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:1fe3f351d51b1d4853f61bc6f58a04a73215ac2c09e5304bb917e62e70a0152b","observation_id":"ddb7adb3-55ea-4e58-8a38-dee92c250240","resolution":{"observed_at":"2026-08-06T23:26:58.004489Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T23:26:57.983603Z","title":"Spatiotemporal co-attention recurrent neural networks for human-skeleton motion prediction","venue":null,"work_id":"b00d881b-b4dd-453f-92f2-7c57445ab150","year":2021},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":27,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.589757Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:23f280811301c3ff1fef1eba6ed6b0413b444f84f1d1fef3469cba57100c24f4","observation_id":"e10cd162-444b-4edc-b00a-b16ec53fb830","resolution":{"observed_at":"2026-08-06T23:26:57.988918Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T23:26:57.968326Z","title":"Multi-granularity anchor-contrastive representation learning for semi-supervised skeleton-based action recognition","venue":null,"work_id":"1f70d099-7e60-4edb-8ac7-86ceaefe995e","year":2022},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":28,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.596748Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:26c84af29613a3ee83112637d71d68016209160bf6969d40065040c8039ac5f6","observation_id":"60f1719e-bca3-491f-90f8-95f0c5214584","resolution":{"observed_at":"2026-08-06T23:26:57.973377Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2302.01647","last_updated":"2024-08-11T15:59:30Z","snapshot_observed_at":"2026-08-13T12:52:09.824745Z","submitted_at":"2023-02-03T10:48:24Z","title":"Blockwise Self-Supervised Learning at Scale","version":2},"cited_work":{"arxiv_id":"2302.01647","doi":null,"metadata_source":"pith","pith_arxiv_id":"2302.01647","snapshot_observed_at":"2026-08-06T23:26:57.726720Z","title":"Blockwise Self-Supervised Learning at Scale","venue":"cs.CV","work_id":"9cb38824-dfec-4dda-9fae-73d4c1c73dac","year":2023},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":29,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.601682Z"},"links":{"cited_paper":"/paper/2302.01647","citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:3be79f3b38713a1d4c0bd2e5505c962572c61e49674df5a08f3080c0f79da628","observation_id":"b9930b06-96f9-4454-a215-2bd38e38a9dc","resolution":{"observed_at":"2026-08-06T23:26:57.734033Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1409.1556","last_updated":"2015-04-10T16:25:04Z","snapshot_observed_at":"2026-08-14T23:20:42.336514Z","submitted_at":"2014-09-04T19:48:04Z","title":"Very Deep Convolutional Networks for Large-Scale Image Recognition","version":6},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1409.1556","snapshot_observed_at":"2026-08-06T23:26:57.606756Z","title":"Very deep convolutional networks for large-scale image recognition","venue":null,"work_id":null,"year":2014},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":30,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.606756Z"},"links":{"cited_paper":"/paper/1409.1556","citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:67ad444b95dc908c00c2ea3a86946d537e501e877bf4c8f2fa8ac711b01b637b","observation_id":"c61c2737-1340-41a6-8a3e-7f38716a5ee3","resolution":{"observed_at":"2026-08-06T23:26:57.606756Z","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":"2026-08-06T23:26:57.611920Z","title":"Going deeper with convolutions","venue":null,"work_id":null,"year":2015},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":31,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.611920Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:5539d9bad0aeb37607114bb988800cc378266703f7a6766d7569e7aff73cc365","observation_id":"9cd33261-1e28-45a8-8abe-f2de59e8647c","resolution":{"observed_at":"2026-08-06T23:26:57.611920Z","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":"2026-08-06T23:26:57.941440Z","title":"Coherence constrained graph lstm for group activity recognition","venue":null,"work_id":"d3ee0142-9b4b-469a-9f6a-8741495aa336","year":2019},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":32,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.617219Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:9a6148ce2e12b6bf12c077a3de91eeaae5c7ce6f21e61cd88fb15c4a637e052b","observation_id":"e6bff137-6afd-471f-b044-cf033e9009fa","resolution":{"observed_at":"2026-08-06T23:26:57.946542Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T23:26:57.923848Z","title":"Branchynet: Fast inference via early exiting from deep neural networks","venue":null,"work_id":"35e1436e-25e0-43e0-a25b-8a1d10088d8c","year":2016},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":33,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.621754Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:ef7de25c9c45a64066cad0ad89f54a30491a272a4d18e6979d45407321a5a189","observation_id":"c3e1a715-e5e5-4e38-926e-26cca7277c3d","resolution":{"observed_at":"2026-08-06T23:26:57.929597Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1907.13625","last_updated":"2020-01-23T09:08:54Z","snapshot_observed_at":"2026-08-14T15:59:23.144746Z","submitted_at":"2019-07-31T17:50:51Z","title":"On Mutual Information Maximization for Representation Learning","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1907.13625","snapshot_observed_at":"2026-08-06T23:26:57.626429Z","title":"On mutual information maximization for representation learning","venue":null,"work_id":null,"year":1907},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":34,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.626429Z"},"links":{"cited_paper":"/paper/1907.13625","citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:36ac52f15f84147d1d2a28e108254da51cb9f67426c37cfd15cbace952ffa207","observation_id":"7339ba12-af84-48d9-b96b-df14e369345b","resolution":{"observed_at":"2026-08-06T23:26:57.626429Z","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":"2026-08-06T23:26:57.904462Z","title":"Decomposing end-to-end backpropagation based on SCPL","venue":null,"work_id":"94a6d8c7-dd7a-4ac7-8a9f-e5da28cb40a8","year":2021},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":35,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.631815Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:73691db87c3a4fb8c9170499b9da0edfe530a4cd168e2bd9981c0fbd37cca982","observation_id":"df8cb2d5-0da2-4f9a-9c18-2ec3e0f0a52f","resolution":{"observed_at":"2026-08-06T23:26:57.912169Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T23:26:57.885737Z","title":"Revisiting locally supervised learning: an alternative to end-to-end training","venue":null,"work_id":"c9dd32d1-feba-4163-8fd7-aef1800cb276","year":2021},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":36,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.637259Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:0998f41e9ff1029d940b791abee11c890bc9ca7c222178679b840d6963406423","observation_id":"e01b6e66-013f-4979-b7ca-20e6d397fc93","resolution":{"observed_at":"2026-08-06T23:26:57.891822Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T23:26:57.869182Z","title":"Associated learning: an alternative to end-to-end backpropagation that works on cnn, rnn, and transformer","venue":null,"work_id":"691e1418-52bd-45ea-9106-89a5711fb8b4","year":2021},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":37,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.643276Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:4a067e814b0fc043ca316a8d2c921da747e522c6885bb8dcc7f673e7812c1708","observation_id":"fd0c88f0-a2e2-447d-9e7c-ebffa07a0316","resolution":{"observed_at":"2026-08-06T23:26:57.874035Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T23:26:57.851409Z","title":"Higcin: Hierarchical graph-based cross inference network for group activity recognition","venue":null,"work_id":"af98611f-c1a1-4460-8040-ca2c467d6a5a","year":2020},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":38,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.648582Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:902969c67b7de315ca1160e25fc36001d42475fdeb74578411151cfe26bb3da6","observation_id":"3eebe8f7-85dd-422c-b572-47c69195dd76","resolution":{"observed_at":"2026-08-06T23:26:57.857346Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T23:26:57.834924Z","title":"Towards interpretable deep local learning with successive gradient reconciliation","venue":null,"work_id":"44ef52a4-c6fe-420b-9ed7-cbb6024993f0","year":2024},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":39,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.652962Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:69dc08da47d73b0430635d2ef178e0a84291eae418cdf5bdef3ac583bdb3ecda","observation_id":"d791d001-0a9f-4450-9f21-9334a589695d","resolution":{"observed_at":"2026-08-06T23:26:57.840799Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T23:26:57.816576Z","title":"Barlow twins: Self-supervised learning via redundancy reduction","venue":null,"work_id":"5112a2d8-5c68-420b-91ce-8f0f2c1fbc31","year":2021},"citing_paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput","version":2},"reference_index":40,"source":"arxiv_source","source_observed_at":"2026-08-06T23:26:57.657461Z"},"links":{"citing_paper":"/paper/2506.18193"},"observation_digest":"sha256:abc4eecf364635f5c4da56161b9cc193fee37b0a3ead75bf5c556db9273fa66c","observation_id":"51fa7bcc-d9b0-467a-8985-0f9edb92ece5","resolution":{"observed_at":"2026-08-06T23:26:57.821790Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}}],"paper":{"arxiv_id":"2506.18193","last_updated":"2025-07-15T14:29:38Z","latest_version":2,"primary_category":"cs.LG","snapshot_observed_at":"2026-08-14T08:09:37.896381Z","submitted_at":"2025-06-22T22:50:06Z","title":"DeInfoReg: A Decoupled Learning Framework for Better Training Throughput"},"reference_resolution":{"displayed":40,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":15,"verified_exact":1,"verified_fuzzy":24},"total_outbound_references":40},"refusal":"A citation records a reference. It does not transfer a finding from one paper to another.","schema":"pith.paper-citation-record.v1","standing_sources":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"thesis":"As of 15 August 2026, this Paper Citation Record lists 40 of 40 outbound references and 0 inbound Pith citation observations for arXiv:2506.18193."}