{"as_of":"2026-08-19T16:45:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:0ae64d8c476a788c62654573ed43c01bdd94d43676f67c200b37540fbfe8f272","coverage":[{"denominator":18,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":18,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-09T21:34:28.423054Z","state":"measured"},{"denominator":20,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":20,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-19T06:32:44.657259+00:00","state":"measured"},{"denominator":2,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":2,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-07T15:21:58.289655Z","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":"2026-07-02T06:06:40.548149Z","state":"measured"}],"external_citation_measurements":[],"inbound":[{"citation":{"cited_paper":{"arxiv_id":"2501.19067","last_updated":"2025-05-21T12:59:55Z","snapshot_observed_at":"2026-08-15T06:06:50.004082Z","submitted_at":"2025-01-31T11:53:16Z","title":"From Low Intrinsic Dimensionality to Non-Vacuous Generalization Bounds in Deep Multi-Task Learning","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2501.19067","snapshot_observed_at":"2026-08-07T15:21:58.289655Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2505.15579","last_updated":"2025-05-21T14:30:59Z","snapshot_observed_at":"2026-08-12T10:44:24.869238Z","submitted_at":"2025-05-21T14:30:59Z","title":"Federated Learning with Unlabeled Clients: Personalization Can Happen in Low Dimensions","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-07T15:21:58.289655Z"},"links":{"cited_paper":"/paper/2501.19067","citing_paper":"/paper/2505.15579"},"observation_digest":"sha256:360d335e499bbb4922ad3047c6beb30acc2fcaac1460d53af668a2c7d0993bcf","observation_id":"4e2b1fda-f970-46d1-b360-1d182824253f","resolution":{"observed_at":"2026-08-07T15:21:58.289655Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2501.19067","last_updated":"2025-05-21T12:59:55Z","snapshot_observed_at":"2026-08-15T06:06:50.004082Z","submitted_at":"2025-01-31T11:53:16Z","title":"From Low Intrinsic Dimensionality to Non-Vacuous Generalization Bounds in Deep Multi-Task Learning","version":2},"cited_work":{"arxiv_id":"2501.19067","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2501.19067","snapshot_observed_at":"2026-07-02T06:06:40.548149Z","title":null,"venue":null,"work_id":"21831a58-9b27-4082-945c-65980a8e23ca","year":2025},"citing_paper":{"arxiv_id":"2607.00995","last_updated":"2026-07-01T14:31:36Z","snapshot_observed_at":"2026-08-19T08:54:11.669769Z","submitted_at":"2026-07-01T14:31:36Z","title":"Deep Multitask Learning for Mixed-Type Outcomes with Shared Sparsity","version":1},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-07-02T06:04:54.352506Z"},"links":{"cited_paper":"/paper/2501.19067","citing_paper":"/paper/2607.00995"},"observation_digest":"sha256:96a44284d449a22ecde3b2a5f9d9e3b587c7a94f14db419fbc26f137d2fe2aaf","observation_id":"0afa58dd-e5d7-4816-819f-cd4373287e03","resolution":{"observed_at":"2026-07-02T06:06:40.550027Z","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"}}],"links":{"evidence":"/evidence","html":"/paper/2501.19067/citation-record","integrity":"/paper/2501.19067/integrity","json":"/paper/2501.19067/citation-record.json","paper":"/paper/2501.19067"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-09T21:34:28.895472Z","title":"As Table 2 shows, the numeric bounds obtained this way can be tighter bound than the upper bound from the Theorem","venue":null,"work_id":"e981c3e9-0a46-46a1-9516-a68d3f4174d3","year":2017},"citing_paper":{"arxiv_id":"2501.19067","last_updated":"2025-05-21T12:59:55Z","snapshot_observed_at":"2026-08-15T06:06:50.004082Z","submitted_at":"2025-01-31T11:53:16Z","title":"From Low Intrinsic Dimensionality to Non-Vacuous Generalization Bounds in Deep Multi-Task Learning","version":2},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-09T21:34:28.385489Z"},"links":{"citing_paper":"/paper/2501.19067"},"observation_digest":"sha256:6ef2c182ce78794fd2179bd9f3f9ebd3aeda4d2d76fcf774726692a982018b9b","observation_id":"ed57d9c3-8aa4-4729-aa50-f4515ca11c23","resolution":{"observed_at":"2026-08-09T21:34:28.899505Z","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":"2026-08-09T21:34:28.883865Z","title":"For each E ∈ E, let lE : S∞ n=1 F n → N be the length function of the multi-task encoder given E","venue":null,"work_id":"5aa9f805-5380-4f99-9be8-637971a64b2f","year":1949},"citing_paper":{"arxiv_id":"2501.19067","last_updated":"2025-05-21T12:59:55Z","snapshot_observed_at":"2026-08-15T06:06:50.004082Z","submitted_at":"2025-01-31T11:53:16Z","title":"From Low Intrinsic Dimensionality to Non-Vacuous Generalization Bounds in Deep Multi-Task Learning","version":2},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-09T21:34:28.389103Z"},"links":{"citing_paper":"/paper/2501.19067"},"observation_digest":"sha256:1222f92923b70d4fdcf76375af0acfe034048bfb29d4f33e10261951dded2ace","observation_id":"c122617c-77ae-4314-8eea-e2010cfb06fa","resolution":{"observed_at":"2026-08-09T21:34:28.887763Z","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":"2026-08-09T21:34:28.907180Z","title":", fn) − R(f1,","venue":null,"work_id":"0802069d-9148-4c5b-90e8-3be8bd502f9b","year":2002},"citing_paper":{"arxiv_id":"2501.19067","last_updated":"2025-05-21T12:59:55Z","snapshot_observed_at":"2026-08-15T06:06:50.004082Z","submitted_at":"2025-01-31T11:53:16Z","title":"From Low Intrinsic Dimensionality to Non-Vacuous Generalization Bounds in Deep Multi-Task Learning","version":2},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-09T21:34:28.381677Z"},"links":{"citing_paper":"/paper/2501.19067"},"observation_digest":"sha256:1e7a359d6beaf477e67ec4be3cfac1a7074171f9716e1203984709f7747b6161","observation_id":"1fad952c-6916-4a31-94fe-f4b29ec71d80","resolution":{"observed_at":"2026-08-09T21:34:28.910937Z","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":"2404.18961","last_updated":"2024-04-29T05:23:10Z","snapshot_observed_at":"2026-08-18T01:32:38.750799Z","submitted_at":"2024-04-29T05:23:10Z","title":"Unleashing the Power of Multi-Task Learning: A Comprehensive Survey Spanning Traditional, Deep, and Pretrained Foundation Model Eras","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2404.18961","snapshot_observed_at":"2026-08-09T21:34:28.373059Z","title":"Unleashing the power of multi-task learning: A comprehensive survey spanning traditional, deep, and pre-trained foundation model eras","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2501.19067","last_updated":"2025-05-21T12:59:55Z","snapshot_observed_at":"2026-08-15T06:06:50.004082Z","submitted_at":"2025-01-31T11:53:16Z","title":"From Low Intrinsic Dimensionality to Non-Vacuous Generalization Bounds in Deep Multi-Task Learning","version":2},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-09T21:34:28.373059Z"},"links":{"cited_paper":"/paper/2404.18961","citing_paper":"/paper/2501.19067"},"observation_digest":"sha256:97c182bc8ddd23dc8d875ec4058248d9c50cb6290164ee2c81e94ee46cf2b3a4","observation_id":"82f123e3-0247-471d-96b6-d323c7dbe9de","resolution":{"observed_at":"2026-08-09T21:34:28.373059Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1806.00582","last_updated":"2022-07-21T12:33:15Z","snapshot_observed_at":"2026-08-15T04:44:38.370440Z","submitted_at":"2018-06-02T04:45:58Z","title":"Federated Learning with Non-IID Data","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1806.00582","snapshot_observed_at":"2026-08-09T21:34:28.377171Z","title":"Federated learning with non-iid data","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2501.19067","last_updated":"2025-05-21T12:59:55Z","snapshot_observed_at":"2026-08-15T06:06:50.004082Z","submitted_at":"2025-01-31T11:53:16Z","title":"From Low Intrinsic Dimensionality to Non-Vacuous Generalization Bounds in Deep Multi-Task Learning","version":2},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-09T21:34:28.377171Z"},"links":{"cited_paper":"/paper/1806.00582","citing_paper":"/paper/2501.19067"},"observation_digest":"sha256:45cb4e0f96b31d6c3ab9a1a3a40782c97b6d339a63c857731079d47bab0c1bd8","observation_id":"c2eba7d1-f75a-402b-9a07-978f00ce36ba","resolution":{"observed_at":"2026-08-09T21:34:28.377171Z","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":"2026-08-09T21:34:28.873122Z","title":null,"venue":null,"work_id":"f7e42b4a-e82d-4782-b8bc-3fe61e806b3b","year":2022},"citing_paper":{"arxiv_id":"2501.19067","last_updated":"2025-05-21T12:59:55Z","snapshot_observed_at":"2026-08-15T06:06:50.004082Z","submitted_at":"2025-01-31T11:53:16Z","title":"From Low Intrinsic Dimensionality to Non-Vacuous Generalization Bounds in Deep Multi-Task Learning","version":2},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-09T21:34:28.392993Z"},"links":{"citing_paper":"/paper/2501.19067"},"observation_digest":"sha256:0072daa08b2d91d267c50208196a432bed495551ae27bb10f8eda8e9ff48db0c","observation_id":"3c2eba39-a27c-4886-a0f8-8c2143a12375","resolution":{"observed_at":"2026-08-09T21:34:28.876451Z","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":"2026-08-09T21:34:28.862827Z","title":"We instantiate (17) with a value tE;F such that 2√mne−tE;F mn = wE;F , i.e","venue":null,"work_id":"ea6ca63e-9e22-4140-83dd-c556fb778488","year":2022},"citing_paper":{"arxiv_id":"2501.19067","last_updated":"2025-05-21T12:59:55Z","snapshot_observed_at":"2026-08-15T06:06:50.004082Z","submitted_at":"2025-01-31T11:53:16Z","title":"From Low Intrinsic Dimensionality to Non-Vacuous Generalization Bounds in Deep Multi-Task Learning","version":2},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-09T21:34:28.396812Z"},"links":{"citing_paper":"/paper/2501.19067"},"observation_digest":"sha256:b4fdf9f9668ae96b5ed24cd7a58cc999ef43d88a8c9fd422f5c8378af08900c5","observation_id":"48745fec-6347-4510-ba76-74b5a5a61e5c","resolution":{"observed_at":"2026-08-09T21:34:28.866337Z","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":"2026-08-09T21:34:28.852264Z","title":null,"venue":null,"work_id":"2ccb65a8-5d26-44f8-af0d-36434b1e198c","year":2010},"citing_paper":{"arxiv_id":"2501.19067","last_updated":"2025-05-21T12:59:55Z","snapshot_observed_at":"2026-08-15T06:06:50.004082Z","submitted_at":"2025-01-31T11:53:16Z","title":"From Low Intrinsic Dimensionality to Non-Vacuous Generalization Bounds in Deep Multi-Task Learning","version":2},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-09T21:34:28.400695Z"},"links":{"citing_paper":"/paper/2501.19067"},"observation_digest":"sha256:213bec7fbb1a1a9477e5f3145498ad7b59725ea6c3a9c0a9e033e59b8452c13c","observation_id":"cfa8e904-7cb0-48a4-963f-e16c4118bd7f","resolution":{"observed_at":"2026-08-09T21:34:28.855509Z","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":"2026-08-09T21:34:28.841777Z","title":"Then we have: E[et kl( Y t | µ t )] ≤ 2 √ t (21) where kl(q|p) = q log q p + (1− q) log1−q 1−p is the Kullback-Leibler divergence between Bernoulli distributions with mean q and p","venue":null,"work_id":"286cd041-254e-46d9-9dac-69022ed54765","year":2004},"citing_paper":{"arxiv_id":"2501.19067","last_updated":"2025-05-21T12:59:55Z","snapshot_observed_at":"2026-08-15T06:06:50.004082Z","submitted_at":"2025-01-31T11:53:16Z","title":"From Low Intrinsic Dimensionality to Non-Vacuous Generalization Bounds in Deep Multi-Task Learning","version":2},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-09T21:34:28.405026Z"},"links":{"citing_paper":"/paper/2501.19067"},"observation_digest":"sha256:97ce8ac43013934c2747c5a46d09c7492367b56948edb851bee72d99b059af38","observation_id":"c17ea713-03b8-4dd5-9f60-a1c3a5fdada3","resolution":{"observed_at":"2026-08-09T21:34:28.845402Z","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":"2026-08-09T21:34:28.830864Z","title":"B.2 Model architectures For the MNIST experiments, we use convolutional networks used in Amit & Meir (2018)","venue":null,"work_id":"c82a1334-7a56-4426-851a-01c548b12c55","year":2018},"citing_paper":{"arxiv_id":"2501.19067","last_updated":"2025-05-21T12:59:55Z","snapshot_observed_at":"2026-08-15T06:06:50.004082Z","submitted_at":"2025-01-31T11:53:16Z","title":"From Low Intrinsic Dimensionality to Non-Vacuous Generalization Bounds in Deep Multi-Task Learning","version":2},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-09T21:34:28.408655Z"},"links":{"citing_paper":"/paper/2501.19067"},"observation_digest":"sha256:ffcea2c5a5d2cf80264587a21faecd0627527e8b005f12463ea87204e141f036","observation_id":"9fe5f07c-fcfa-4f44-b05b-36df577acd0d","resolution":{"observed_at":"2026-08-09T21:34:28.834630Z","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":"2026-08-09T21:34:28.819173Z","title":"(2024), and a ViT model pretrained from ImageNet (Dosovitskiy et al., 2021)","venue":null,"work_id":"cb9c55e2-59b8-4255-b624-42b23ac29797","year":2024},"citing_paper":{"arxiv_id":"2501.19067","last_updated":"2025-05-21T12:59:55Z","snapshot_observed_at":"2026-08-15T06:06:50.004082Z","submitted_at":"2025-01-31T11:53:16Z","title":"From Low Intrinsic Dimensionality to Non-Vacuous Generalization Bounds in Deep Multi-Task Learning","version":2},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-09T21:34:28.412278Z"},"links":{"citing_paper":"/paper/2501.19067"},"observation_digest":"sha256:c25fbca59b429ab0bbb097eeb72c3304171e25132f56c4f1ff7c7e7d2464707d","observation_id":"5d490a25-4faf-4ec1-9d77-ac00ad10ec93","resolution":{"observed_at":"2026-08-09T21:34:28.822816Z","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":"2026-08-09T21:34:28.808241Z","title":"far more than the available number of samples per task","venue":null,"work_id":"49780e64-1199-4629-9f39-a7a652eda035","year":2022},"citing_paper":{"arxiv_id":"2501.19067","last_updated":"2025-05-21T12:59:55Z","snapshot_observed_at":"2026-08-15T06:06:50.004082Z","submitted_at":"2025-01-31T11:53:16Z","title":"From Low Intrinsic Dimensionality to Non-Vacuous Generalization Bounds in Deep Multi-Task Learning","version":2},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-09T21:34:28.415962Z"},"links":{"citing_paper":"/paper/2501.19067"},"observation_digest":"sha256:879810f0fc834a8e37273e23dd36431f9cab5e2efe2035712e8bdf222ec3b475","observation_id":"0e1fd733-c4ca-4af4-8b40-c9ce2c7a6f38","resolution":{"observed_at":"2026-08-09T21:34:28.811711Z","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":"2026-08-09T21:34:28.796523Z","title":"By this construction, the matrix P ∈ RD×d never has to be explicitly instantiated, which makes the memory and computational overhead tractable","venue":null,"work_id":"807d1b1c-d3e3-48c2-a165-15dd94de0cae","year":2000},"citing_paper":{"arxiv_id":"2501.19067","last_updated":"2025-05-21T12:59:55Z","snapshot_observed_at":"2026-08-15T06:06:50.004082Z","submitted_at":"2025-01-31T11:53:16Z","title":"From Low Intrinsic Dimensionality to Non-Vacuous Generalization Bounds in Deep Multi-Task Learning","version":2},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-09T21:34:28.419391Z"},"links":{"citing_paper":"/paper/2501.19067"},"observation_digest":"sha256:57138de6e49fdb7378e22322f5015171822c2813f21ecbf836d7bfbcc7e52d9d","observation_id":"8c0c2c59-b44c-4795-9b8a-30c08db12e6d","resolution":{"observed_at":"2026-08-09T21:34:28.800763Z","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":"2026-08-09T21:34:28.785882Z","title":null,"venue":null,"work_id":"3e1becb3-8083-4367-9a49-80bc5107053f","year":2000},"citing_paper":{"arxiv_id":"2501.19067","last_updated":"2025-05-21T12:59:55Z","snapshot_observed_at":"2026-08-15T06:06:50.004082Z","submitted_at":"2025-01-31T11:53:16Z","title":"From Low Intrinsic Dimensionality to Non-Vacuous Generalization Bounds in Deep Multi-Task Learning","version":2},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-09T21:34:28.423054Z"},"links":{"citing_paper":"/paper/2501.19067"},"observation_digest":"sha256:c85c345c9c5bf7777ff7909c93c75829d4858869e1e42b4abf101a3d5c83cc3f","observation_id":"b2406062-db33-4bb4-8b16-d3e6ea888e86","resolution":{"observed_at":"2026-08-09T21:34:28.789301Z","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":"2310.11453","last_updated":"2023-10-17T17:59:15Z","snapshot_observed_at":"2026-08-15T04:43:14.762539Z","submitted_at":"2023-10-17T17:59:15Z","title":"BitNet: Scaling 1-bit Transformers for Large Language Models","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2310.11453","snapshot_observed_at":"2026-08-09T21:34:28.369362Z","title":"BitNet: Scaling 1-bit transformers for large language models","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2501.19067","last_updated":"2025-05-21T12:59:55Z","snapshot_observed_at":"2026-08-15T06:06:50.004082Z","submitted_at":"2025-01-31T11:53:16Z","title":"From Low Intrinsic Dimensionality to Non-Vacuous Generalization Bounds in Deep Multi-Task Learning","version":2},"reference_index":1971,"source":"pdf_text","source_observed_at":"2026-08-09T21:34:28.369362Z"},"links":{"cited_paper":"/paper/2310.11453","citing_paper":"/paper/2501.19067"},"observation_digest":"sha256:30d9ad903371e476d40363142c3b691b2d0dbc59dcf0a4e380223135243f996f","observation_id":"a5175669-0060-4dd6-a9c9-57ea1796f11c","resolution":{"observed_at":"2026-08-09T21:34:28.369362Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2102.03748","last_updated":"2021-02-07T09:03:43Z","snapshot_observed_at":"2026-08-16T18:47:14.759078Z","submitted_at":"2021-02-07T09:03:43Z","title":"PAC-Bayes Bounds for Meta-learning with Data-Dependent Prior","version":1},"cited_work":{"arxiv_id":"2102.03748","doi":null,"metadata_source":"pith","pith_arxiv_id":"2102.03748","snapshot_observed_at":"2026-08-09T21:34:28.774240Z","title":"PAC-Bayes Bounds for Meta-learning with Data-Dependent Prior","venue":"cs.LG","work_id":"b73f3c2b-931f-456d-afcd-a1f52e48aacf","year":2021},"citing_paper":{"arxiv_id":"2501.19067","last_updated":"2025-05-21T12:59:55Z","snapshot_observed_at":"2026-08-15T06:06:50.004082Z","submitted_at":"2025-01-31T11:53:16Z","title":"From Low Intrinsic Dimensionality to Non-Vacuous Generalization Bounds in Deep Multi-Task Learning","version":2},"reference_index":2018,"source":"pdf_text","source_observed_at":"2026-08-09T21:34:28.357918Z"},"links":{"cited_paper":"/paper/2102.03748","citing_paper":"/paper/2501.19067"},"observation_digest":"sha256:aabf8fce523343199dc63f2139a7da5596b27cb045b3347a80c93788eaddfa62","observation_id":"3b1ca76b-601b-41b5-819d-26e098338e6e","resolution":{"observed_at":"2026-08-09T21:34:28.778131Z","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":{"arxiv_id":"2302.11709","last_updated":"2023-02-23T00:07:03Z","snapshot_observed_at":"2026-08-16T15:53:23.819428Z","submitted_at":"2023-02-23T00:07:03Z","title":"Bayes meets Bernstein at the Meta Level: an Analysis of Fast Rates in Meta-Learning with PAC-Bayes","version":1},"cited_work":{"arxiv_id":"2302.11709","doi":null,"metadata_source":"pith","pith_arxiv_id":"2302.11709","snapshot_observed_at":"2026-08-09T21:34:28.488682Z","title":"Bayes meets Bernstein at the Meta Level: an Analysis of Fast Rates in Meta-Learning with PAC-Bayes","venue":"stat.ML","work_id":"d15ee598-f7db-413b-822d-e18337ce6688","year":2023},"citing_paper":{"arxiv_id":"2501.19067","last_updated":"2025-05-21T12:59:55Z","snapshot_observed_at":"2026-08-15T06:06:50.004082Z","submitted_at":"2025-01-31T11:53:16Z","title":"From Low Intrinsic Dimensionality to Non-Vacuous Generalization Bounds in Deep Multi-Task Learning","version":2},"reference_index":2022,"source":"pdf_text","source_observed_at":"2026-08-09T21:34:28.365546Z"},"links":{"cited_paper":"/paper/2302.11709","citing_paper":"/paper/2501.19067"},"observation_digest":"sha256:943b94ca7a66e219985c254c7307b01ed453f32cd763e61860fb7b2f910bb503","observation_id":"eb086567-092c-41ce-8636-f21dd08e6ba2","resolution":{"observed_at":"2026-08-09T21:34:28.495200Z","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":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-09T21:34:28.757362Z","title":"A note on the PAC Bayesian theorem","venue":null,"work_id":"9f3eb225-f83a-45fe-8566-9f6753775bd1","year":null},"citing_paper":{"arxiv_id":"2501.19067","last_updated":"2025-05-21T12:59:55Z","snapshot_observed_at":"2026-08-15T06:06:50.004082Z","submitted_at":"2025-01-31T11:53:16Z","title":"From Low Intrinsic Dimensionality to Non-Vacuous Generalization Bounds in Deep Multi-Task Learning","version":2},"reference_index":2024,"source":"pdf_text","source_observed_at":"2026-08-09T21:34:28.362164Z"},"links":{"citing_paper":"/paper/2501.19067"},"observation_digest":"sha256:8cd6575669afa3fead44e111f811709bf5425e2d4c96919687648315e29060c3","observation_id":"d9e06bbf-441a-4089-85c2-7f22347f3ad9","resolution":{"observed_at":"2026-08-09T21:34:28.760986Z","resolver_source":"arxiv_id_nonexistent","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"}}],"paper":{"arxiv_id":"2501.19067","last_updated":"2025-05-21T12:59:55Z","latest_version":2,"primary_category":"cs.LG","snapshot_observed_at":"2026-08-15T06:06:50.004082Z","submitted_at":"2025-01-31T11:53:16Z","title":"From Low Intrinsic Dimensionality to Non-Vacuous Generalization Bounds in Deep Multi-Task Learning"},"reference_resolution":{"displayed":18,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":6,"verified_exact":3,"verified_fuzzy":9},"total_outbound_references":18},"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 18 of 18 outbound references and 2 inbound Pith citation observations for arXiv:2501.19067."}