{"as_of":"2026-08-17T04:14:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:724fcd6c0b833fab2fc471e29cc4ef2536b8d37c0d8ede6fc48b3e63788dbfbf","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":"2026-08-15T18:39:11.512973Z","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":"2026-08-16T06:30:59.297886+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/2506.19836/citation-record","integrity":"/paper/2506.19836/integrity","json":"/paper/2506.19836/citation-record.json","paper":"/paper/2506.19836"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T18:39:11.337641Z","title":"Deep learning with differential privacy","venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.337641Z"},"links":{"citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:30aa9fd25912031270685c95eda8a1c056516ea393ec97007c3c707b2b649e9d","observation_id":"b63d38ef-3642-41ca-a18d-60b490cac01d","resolution":{"observed_at":"2026-08-15T18:39:11.337641Z","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":"2026-08-15T18:39:12.046077Z","title":"Pri- vacy amplification by subsampling: Tight analyses via couplings and divergences","venue":null,"work_id":"47b40cc4-4e9d-4453-896f-6fb2077c9c49","year":2018},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.343801Z"},"links":{"citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:62b3ca3c98d3f7925e3b61e327080a9730efc2e582e610969ceb0d4b366a12e6","observation_id":"4e3ff84b-d9dc-4760-a0f7-a1807bfbb3c3","resolution":{"observed_at":"2026-08-15T18:39:12.050746Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T18:39:12.033349Z","title":"Private stochastic convex optimization with optimal rates","venue":null,"work_id":"9d0a9f1f-0f92-4323-8863-58b9f1bc2313","year":2019},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.349474Z"},"links":{"citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:7da0a5cb3be7b2bfeeb2f186f355b3406e57e534ca528938e1eda5bc77465df3","observation_id":"2343b987-2662-4fe2-bcf6-9698aac5ade3","resolution":{"observed_at":"2026-08-15T18:39:12.037515Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T18:39:12.019381Z","title":"Broadening the scope of differential privacy using metrics","venue":null,"work_id":"dac607fb-ac6b-451d-8ffa-64230122812f","year":2013},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.354676Z"},"links":{"citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:f5036f1fc88f301b5fcae6ea4113cd860fe8ce88b04d4762164311b6015ddfb3","observation_id":"ba4d3bdf-096f-4f72-b27b-b4b886ca87e1","resolution":{"observed_at":"2026-08-15T18:39:12.024056Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T18:39:12.005983Z","title":"On the relationships between no- tions of simulation-based security","venue":null,"work_id":"46927b51-3977-48ac-bc1a-ef3bd9115183","year":2005},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.358806Z"},"links":{"citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:f47cc370b96204abe5b45c06bb298864157eb371b9fba37036cd64659910dc39","observation_id":"496e2c7f-fd8f-435e-a297-5d3c9abf517d","resolution":{"observed_at":"2026-08-15T18:39:12.010840Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2204.13650","last_updated":"2022-06-16T17:47:42Z","snapshot_observed_at":"2026-08-16T17:03:52.960553Z","submitted_at":"2022-04-28T17:10:56Z","title":"Unlocking High-Accuracy Differentially Private Image Classification through Scale","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2204.13650","snapshot_observed_at":"2026-08-15T18:39:11.363771Z","title":"Unlocking high-accuracy differentially private image classification through scale","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.363771Z"},"links":{"cited_paper":"/paper/2204.13650","citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:d8e0f0274e016ff7a7736a1ff0a38acf294c3993878e9ae1dc2730df9c8b357d","observation_id":"9fe58ce3-6247-4293-a0dd-ae3c6d2379d5","resolution":{"observed_at":"2026-08-15T18:39:11.363771Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1905.02383","last_updated":"2019-05-30T23:51:04Z","snapshot_observed_at":"2026-08-17T01:44:36.763458Z","submitted_at":"2019-05-07T06:57:19Z","title":"Gaussian Differential Privacy","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1905.02383","snapshot_observed_at":"2026-08-15T18:39:11.368659Z","title":"Gaussian differential privacy","venue":null,"work_id":null,"year":1905},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.368659Z"},"links":{"cited_paper":"/paper/1905.02383","citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:b588487d126d8e2403efea668607187ff48f5a391db1b6f0898df70b03ecfd97","observation_id":"012a1038-9c87-4aec-a137-8786b9e7d69e","resolution":{"observed_at":"2026-08-15T18:39:11.368659Z","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":"2026-08-15T18:39:11.993792Z","title":"Differential privacy","venue":null,"work_id":"88756d35-8e85-46e5-bd1b-4b12c3d78240","year":2006},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.373419Z"},"links":{"citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:18688efc18bcc7293f1bf05bec2380faf398f6ff8be8e176be2315b8af685cc5","observation_id":"fee73f2d-a68d-49c0-b593-55bae5e389c7","resolution":{"observed_at":"2026-08-15T18:39:11.997615Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T18:39:11.980756Z","title":"Why is public pretraining necessary for private model training? In Interna- tional Conference on Machine Learning, pages 10611– 10627","venue":null,"work_id":"3a2314b0-a044-4eaf-b276-06000318f3ca","year":2023},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.378450Z"},"links":{"citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:00a526161a5d2579aa50a8af98657053f0a0932c613ecab2e3bd30e0e0d0a165","observation_id":"f68d1b85-816a-408a-9aa0-049616cff5a1","resolution":{"observed_at":"2026-08-15T18:39:11.985514Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T18:39:11.382625Z","title":"Property inference attacks on fully connected neural networks using permutation invariant representations","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.382625Z"},"links":{"citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:d5a28b4e2b1c727335f1f786a2bcb2e46d8e25059dc3ea8ef704002846166ddb","observation_id":"2da74714-39dd-4cb3-a608-f312f33721d0","resolution":{"observed_at":"2026-08-15T18:39:11.382625Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2302.13861","last_updated":"2023-02-27T15:02:04Z","snapshot_observed_at":"2026-08-16T15:52:24.331573Z","submitted_at":"2023-02-27T15:02:04Z","title":"Differentially Private Diffusion Models Generate Useful Synthetic Images","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2302.13861","snapshot_observed_at":"2026-08-15T18:39:11.386421Z","title":"Dif- ferentially private diffusion models generate useful synthetic images","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.386421Z"},"links":{"cited_paper":"/paper/2302.13861","citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:d0e3d5a4b36f50e3d853c3d131fd18f42b7f8968a8788a6b5fedd9e185bb9a14","observation_id":"bfb60799-591a-478f-8eb8-6f5e917d2c22","resolution":{"observed_at":"2026-08-15T18:39:11.386421Z","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":"2026-08-15T18:39:11.961043Z","title":"Deep learning with label differential privacy","venue":null,"work_id":"772d4e1d-ff04-48d3-b443-b933c980b312","year":2021},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.391286Z"},"links":{"citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:ade4a19d493929eb6385a1d5ac04aa10506f86c6f03d0d9e997236c2bdee16e6","observation_id":"d7fcf191-7c49-4232-ad5b-e92e20d79262","resolution":{"observed_at":"2026-08-15T18:39:11.964984Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2209.04053","last_updated":"2022-09-08T22:43:50Z","snapshot_observed_at":"2026-08-16T16:33:46.078714Z","submitted_at":"2022-09-08T22:43:50Z","title":"Algorithms with More Granular Differential Privacy Guarantees","version":1},"cited_work":{"arxiv_id":"2209.04053","doi":null,"metadata_source":"pith","pith_arxiv_id":"2209.04053","snapshot_observed_at":"2026-08-15T18:39:11.649150Z","title":"Algorithms with More Granular Differential Privacy Guarantees","venue":"cs.CR","work_id":"c23b2268-3c7c-4abf-90c3-dd9ed0809a8b","year":2022},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.395020Z"},"links":{"cited_paper":"/paper/2209.04053","citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:8ca9eec02ef0c7ef861871df784e01076e4bca2d664b1d46edf1607fa3f3717e","observation_id":"899d6dc7-909c-4fec-8823-10cc9d8fd7e0","resolution":{"observed_at":"2026-08-15T18:39:11.653870Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1603.01508","last_updated":"2017-05-23T18:52:06Z","snapshot_observed_at":"2026-08-14T22:07:28.907198Z","submitted_at":"2016-03-04T15:50:24Z","title":"Inferential Privacy Guarantees for Differentially Private Mechanisms","version":2},"cited_work":{"arxiv_id":"1603.01508","doi":null,"metadata_source":"pith","pith_arxiv_id":"1603.01508","snapshot_observed_at":"2026-08-15T18:39:11.628892Z","title":"Inferential Privacy Guarantees for Differentially Private Mechanisms","venue":"cs.DS","work_id":"90824364-666a-43a8-92a6-05349eb2cfd6","year":2016},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.399322Z"},"links":{"cited_paper":"/paper/1603.01508","citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:da870cb240b88068ab84594eb8f6903cf781b15ebce5516a9609d4b8fe2df141","observation_id":"d24899e4-0cd5-454e-b079-bba7d6305522","resolution":{"observed_at":"2026-08-15T18:39:11.633989Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T18:39:11.947567Z","title":"Numerical composition of differential privacy","venue":null,"work_id":"42e80ebd-537c-4fd9-b353-4864fab1ea91","year":2021},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.404314Z"},"links":{"citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:089a5f1889c87e30d9729b992738af06be517a664d84c150d7e4c93648e6b2e6","observation_id":"938f6768-cf40-4aaa-8d3f-ada0f273b2ab","resolution":{"observed_at":"2026-08-15T18:39:11.952575Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T18:39:11.932863Z","title":"Bounding training data re- construction in private (deep) learning","venue":null,"work_id":"9b30a0fd-cc8b-40b8-929a-63388161ea19","year":2022},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.408398Z"},"links":{"citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:b5e8157bcc2033e41706ad18cfc6abe9c79b988baf94f64ae88bc2fff3ec2a6f","observation_id":"4916a56c-dea4-4305-bd50-d5b357a6dcf7","resolution":{"observed_at":"2026-08-15T18:39:11.938496Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T18:39:11.918649Z","title":"Bounding training data reconstruction in dp-sgd","venue":null,"work_id":"825efef6-d293-4ffa-b3ae-7454a2ef2974","year":2024},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.412063Z"},"links":{"citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:5d734a4a991ae84c27e5187f51eab4b6a8f80d177245a96c27f4ab9ce827f3e4","observation_id":"043e498c-1b50-4783-bf5c-bb06543fbb8d","resolution":{"observed_at":"2026-08-15T18:39:11.923478Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T18:39:11.902500Z","title":"Are attribute inference attacks just imputation? In Proceedings of the 2022 ACM SIGSAC Conference on Computer and Communications Security, pages 1569–1582, 2022","venue":null,"work_id":"523233b6-f56c-4d36-8244-3743911e4ff6","year":2022},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.415655Z"},"links":{"citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:390a3d83646f4ec09b0b7d03adbe7086e50c6fa15d98551c397d24eff94223db","observation_id":"07ef4124-7e6e-47bc-85bf-5def289bcded","resolution":{"observed_at":"2026-08-15T18:39:11.907490Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T18:39:11.890096Z","title":"{AttriGuard}: A practical defense against attribute inference attacks via adversarial machine learning","venue":null,"work_id":"084dfe73-01b2-4aae-adef-6e3e9971d798","year":2018},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.419982Z"},"links":{"citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:7da5297c9c47f47ba1247c6754bcae64ad5e66b6e9ff39ba3c1684ff5f23bb3b","observation_id":"d3ae3943-1dc9-493a-84cb-23a47c7dc65e","resolution":{"observed_at":"2026-08-15T18:39:11.894415Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T18:39:11.878223Z","title":"Private convex empirical risk minimization and high- dimensional regression","venue":null,"work_id":"e70f8aec-889c-461f-a6d2-a14128b1150a","year":2012},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.423896Z"},"links":{"citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:75307ebe8abe11db45158163a65e19950142d84b9a700d9d573eee96f1bb6789","observation_id":"6729356e-bd18-413e-8e1b-639810070269","resolution":{"observed_at":"2026-08-15T18:39:11.882699Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2310.15454","last_updated":"2023-10-24T01:59:28Z","snapshot_observed_at":"2026-08-16T14:49:31.541023Z","submitted_at":"2023-10-24T01:59:28Z","title":"Private Learning with Public Features","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2310.15454","snapshot_observed_at":"2026-08-15T18:39:11.427865Z","title":"Private learning with public features","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.427865Z"},"links":{"cited_paper":"/paper/2310.15454","citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:797534136e6de60763c345b05a385a8e44cf24a69faf58ba84c9dd30d7b64319","observation_id":"54e77845-7a2c-4569-bd4c-15fda8eefb36","resolution":{"observed_at":"2026-08-15T18:39:11.427865Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.05679","last_updated":"2022-11-10T18:42:34Z","snapshot_observed_at":"2026-08-16T17:50:07.783071Z","submitted_at":"2021-10-12T01:45:27Z","title":"Large Language Models Can Be Strong Differentially Private Learners","version":6},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.05679","snapshot_observed_at":"2026-08-15T18:39:11.431975Z","title":"Large language models can be strong differentially private learners","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.431975Z"},"links":{"cited_paper":"/paper/2110.05679","citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:5b63e94532ecafa2c2e82f8dada32015b6b8564eb27a7b9994c524d3ada10007","observation_id":"7485d049-53f0-4fa0-bfb6-2070bcad2ab6","resolution":{"observed_at":"2026-08-15T18:39:11.431975Z","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":"2026-08-15T18:39:11.866847Z","title":"Distributional privacy for data sharing","venue":null,"work_id":"ca96a946-7626-41ce-b529-8edf029d3c18","year":2022},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.437428Z"},"links":{"citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:511b6c0fe22f3023d65b003ab5623ba5c3c16ab6677c4f2dbcbd16f505419d14","observation_id":"95e102cb-06fb-44cc-b02b-efb3d54ea117","resolution":{"observed_at":"2026-08-15T18:39:11.870781Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T18:39:11.854994Z","title":"Property inference from poisoning","venue":null,"work_id":"f89bdf7b-3177-4ecd-83b7-c4a08ec2d094","year":2022},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.442833Z"},"links":{"citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:61ef12efce2e31f7040b11eb00ed9909fb29a5424e2fb73eecbbc2bd01248d51","observation_id":"19817a37-a71e-4d66-a5a9-06d946c806b2","resolution":{"observed_at":"2026-08-15T18:39:11.859140Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T18:39:11.843172Z","title":"Antipodes of label differential privacy: Pate and alibi","venue":null,"work_id":"27340550-9234-474e-9533-073f4ea9f994","year":2021},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.447190Z"},"links":{"citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:8b800c8b8c3580bba0e4b53284b80b8bd9fe5c9adf2a4d299542e6856cc862a7","observation_id":"8593507f-7a4c-4ace-b400-87e10e0b6f10","resolution":{"observed_at":"2026-08-15T18:39:11.847445Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T18:39:11.831770Z","title":"Not all features are equal: Discovering essential features for preserving prediction privacy","venue":null,"work_id":"578a5af0-b979-4b81-b3c8-4e50c94eb24d","year":2021},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.451231Z"},"links":{"citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:ea0277951c3628da2ba477bbd285d9c5f8965f4a47c051cd788200127fe56be0","observation_id":"4b136916-2d23-4f1b-ab65-47ec9478b2e6","resolution":{"observed_at":"2026-08-15T18:39:11.835994Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T18:39:11.817759Z","title":"Stochastic gradient descent for non-smooth optimization: Convergence re- sults and optimal averaging schemes","venue":null,"work_id":"6ad2ed19-81be-4965-964e-e10b4379b91c","year":2013},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.455574Z"},"links":{"citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:8fde92b0c524b186ae4ae0e971ff90d00d43a7a4cd140eb001de5a2b9fa15b64","observation_id":"0c09a8b2-e2ac-4663-bc31-3ee5bdc282e5","resolution":{"observed_at":"2026-08-15T18:39:11.823287Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2108.12944","last_updated":"2022-07-16T22:37:41Z","snapshot_observed_at":"2026-08-16T18:00:13.141255Z","submitted_at":"2021-08-30T01:11:10Z","title":"Selective Differential Privacy for Language Modeling","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2108.12944","snapshot_observed_at":"2026-08-15T18:39:11.460192Z","title":"Selective differential privacy for language model- ing","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.460192Z"},"links":{"cited_paper":"/paper/2108.12944","citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:7e28c814dbb217f2a5327f90b448d7d0145d693cb40fd713ef3ae14806b1dddf","observation_id":"02377e32-d5e9-4847-9dcd-688601cc2949","resolution":{"observed_at":"2026-08-15T18:39:11.460192Z","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":"2026-08-15T18:39:11.804559Z","title":"Stochastic gradient descent with differentially private updates","venue":null,"work_id":"8177ced1-d1f8-43d4-914a-c2a06c5e874b","year":2013},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.464655Z"},"links":{"citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:0bfcd117cd62c7e3736e4e7575dc7a203f8620dc539394d6a8a340c9758a7e06","observation_id":"c9c06199-39e5-4b68-9502-28ee5a0db4c5","resolution":{"observed_at":"2026-08-15T18:39:11.809122Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T18:39:11.790605Z","title":"Machine learning with differentially private labels: Mechanisms and frameworks","venue":null,"work_id":"32a19fcd-1f18-4f9f-9663-a9e28f0e7785","year":2022},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.469005Z"},"links":{"citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:a504962c942a81ac4bfcbd5d89f7d3987570539b165f60e5a5260f5bd62e005b","observation_id":"7923fdc2-f25d-4835-8ce7-02084577d7d7","resolution":{"observed_at":"2026-08-15T18:39:11.795112Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2206.04236","last_updated":"2026-04-07T02:57:51Z","snapshot_observed_at":"2026-07-06T13:18:52.367719Z","submitted_at":"2022-06-09T02:52:57Z","title":"Edgeworth Accountant: An Analytical Approach to Differential Privacy Composition","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2206.04236","snapshot_observed_at":"2026-08-15T18:39:11.472955Z","title":"Analytical composition of differential privacy via the edgeworth accountant","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.472955Z"},"links":{"cited_paper":"/paper/2206.04236","citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:c3d61647139d0345e7ddb23c5444287b2578c88227b2e22479ee74e45e689698","observation_id":"9bb085cd-7b5c-4961-83b5-01c67466fc77","resolution":{"observed_at":"2026-08-15T18:39:11.472955Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2304.07927","last_updated":"2023-11-21T02:15:33Z","snapshot_observed_at":"2026-08-16T15:40:03.889088Z","submitted_at":"2023-04-17T00:38:01Z","title":"A Randomized Approach for Tight Privacy Accounting","version":2},"cited_work":{"arxiv_id":"2304.07927","doi":null,"metadata_source":"pith","pith_arxiv_id":"2304.07927","snapshot_observed_at":"2026-08-15T18:39:11.558974Z","title":"A Randomized Approach for Tight Privacy Accounting","venue":"cs.CR","work_id":"6d2f9d96-6ea8-4855-9941-7896b01f3be5","year":2023},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.477351Z"},"links":{"cited_paper":"/paper/2304.07927","citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:559ad86ba1ac231d6b8697a8f74f50f2379cd25fe5439650fe1788a4e8c2164b","observation_id":"bd6c1182-7dfd-41df-b91a-d8eef087273b","resolution":{"observed_at":"2026-08-15T18:39:11.566090Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T18:39:11.777466Z","title":"A study of face obfuscation in imagenet","venue":null,"work_id":"6f26edd6-1756-48e0-9164-2a91d3b265c2","year":2022},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.481846Z"},"links":{"citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:35facb1a7eece37c1c8a858bc8a7674f832ceaababdebb88b919250b60a02c61","observation_id":"14b26a9b-c696-483e-8211-fbaac99cc077","resolution":{"observed_at":"2026-08-15T18:39:11.782555Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2306.08842","last_updated":"2023-06-28T22:24:33Z","snapshot_observed_at":"2026-08-16T15:23:48.265185Z","submitted_at":"2023-06-15T04:06:24Z","title":"ViP: A Differentially Private Foundation Model for Computer Vision","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2306.08842","snapshot_observed_at":"2026-08-15T18:39:11.486572Z","title":"Vip: A differentially private foundation model for computer vision","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.486572Z"},"links":{"cited_paper":"/paper/2306.08842","citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:9abc9f085d8509b4df4d4e1736929199cff91c82bbacff80f27784b7db9d2728","observation_id":"dbd68a31-c612-4b31-8af6-aeee9d488568","resolution":{"observed_at":"2026-08-15T18:39:11.486572Z","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":"2026-08-15T18:39:11.763860Z","title":"Attribute privacy: Framework and mechanisms","venue":null,"work_id":"d1e70018-cf6b-4c6f-8963-5a8e40105917","year":2022},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.491183Z"},"links":{"citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:ed35d3842790a5d9ff3bea6fc2c94efa94e3608553cc0b440774b36ccb223c23","observation_id":"8cd2078a-c2c9-4d5f-9332-e734bc98f034","resolution":{"observed_at":"2026-08-15T18:39:11.768361Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T18:39:11.750589Z","title":"Opti- mal accounting of differential privacy via characteristic function","venue":null,"work_id":"d8345345-461f-49c2-998a-d68e048fac15","year":2022},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.495767Z"},"links":{"citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:7fa0003a3da7f2c96c7450b7d30cb64512abcd92ba347582ab9de72128b17c1c","observation_id":"f3e3ce7a-31fc-4c82-958f-fb5004d87985","resolution":{"observed_at":"2026-08-15T18:39:11.754881Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T18:39:11.737815Z","title":"Feature Differential Privacy","venue":null,"work_id":"98e9a684-1503-4f18-bd32-d46f64e9de91","year":null},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.500435Z"},"links":{"citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:b99dd46e8e3480703bb52b49679ad8258c51688b34f5aeefc71c9a4e416afc88","observation_id":"e7a0447a-4af2-4e7d-b02e-d12e4511c4c1","resolution":{"observed_at":"2026-08-15T18:39:11.742048Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T18:39:11.725330Z","title":null,"venue":null,"work_id":"a617f8c3-6fe9-4d2b-87da-035c21406f76","year":null},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.504589Z"},"links":{"citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:338bcf4ed57db7a2acfadefc464275a3633fff7b95009ca13d3c06211eb41a66","observation_id":"1f680f1a-be39-461c-a67d-8d517616f59d","resolution":{"observed_at":"2026-08-15T18:39:11.729477Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T18:39:11.712023Z","title":null,"venue":null,"work_id":"6774367e-4238-4586-ae7b-2bb631527a04","year":null},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.508859Z"},"links":{"citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:d8db4a9a42b01bbbbcad049c5c3f63ef87f90b6ae07a0116376590b937cfd979","observation_id":"2710cf82-4489-4dac-9a04-96482b802bb7","resolution":{"observed_at":"2026-08-15T18:39:11.717107Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T18:39:11.699183Z","title":null,"venue":null,"work_id":"dfcbf5f5-c1e4-48d3-8f57-6c8d4ff4a094","year":null},"citing_paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes","version":1},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-08-15T18:39:11.512973Z"},"links":{"citing_paper":"/paper/2506.19836"},"observation_digest":"sha256:876014a869aade039aede1818c7d515e7ff6addafb7db7c0ec855313812dc5fe","observation_id":"5137210c-bf0e-4d33-9570-87440b29726a","resolution":{"observed_at":"2026-08-15T18:39:11.703540Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}}],"paper":{"arxiv_id":"2506.19836","last_updated":"2025-06-24T17:53:28Z","latest_version":1,"primary_category":"cs.CR","snapshot_observed_at":"2026-08-16T21:13:19.653743Z","submitted_at":"2025-06-24T17:53:28Z","title":"Machine Learning with Privacy for Protected Attributes"},"reference_resolution":{"displayed":40,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":13,"verified_exact":3,"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":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"thesis":"As of 17 August 2026, this Paper Citation Record lists 40 of 40 outbound references and 0 inbound Pith citation observations for arXiv:2506.19836."}