{"as_of":"2026-08-23T01:21:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:fb304f718bb6174b6ddacfcdd2489b1fa9f5cb80bcae85560ad03c92091a64d3","coverage":[{"denominator":0,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":0,"source":"paper_references, paper_reference_links","source_observed_at":null,"state":"measured"},{"denominator":71,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":71,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-22T06:32:14.747728+00:00","state":"measured"},{"denominator":71,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":71,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-15T23:22:16.551840Z","state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":1,"source":"pith","source_observed_at":"2026-08-05T02:28:24.338817Z","state":"measured"}],"external_citation_measurements":[{"count":377,"observed_at":"2026-08-05T02:28:24.338817Z","source":"pith"}],"inbound":[{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":"2001.02610","doi":"10.48550/arxiv.2001.02610","metadata_source":"pith","pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"iDLG: Improved deep leakage from gradients","venue":"cs.LG","work_id":"611e14b3-8402-48e3-92f8-d9356cbdc89d","year":2020},"citing_paper":{"arxiv_id":"2308.06822","last_updated":"2026-05-15T12:55:02Z","snapshot_observed_at":"2026-08-19T02:45:58.022330Z","submitted_at":"2023-08-13T17:40:56Z","title":"Approximate and Weighted Data Reconstruction Attack in Federated Learning","version":3},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-05-24T07:20:31.753349Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2308.06822"},"observation_digest":"sha256:47d935b536762e54139e78d14469e46ca719e5252020b9c3aca5dd9cc0ca8cd9","observation_id":"aa6bd328-f1c1-4768-9d5c-19023c010712","resolution":{"observed_at":"2026-05-24T07:24:06.143857Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":"2001.02610","doi":"10.48550/arxiv.2001.02610","metadata_source":"pith","pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"iDLG: Improved deep leakage from gradients","venue":"cs.LG","work_id":"611e14b3-8402-48e3-92f8-d9356cbdc89d","year":2020},"citing_paper":{"arxiv_id":"2406.10861","last_updated":"2024-06-16T09:12:16Z","snapshot_observed_at":"2026-08-13T22:47:40.173669Z","submitted_at":"2024-06-16T09:12:16Z","title":"Knowledge Distillation in Federated Learning: a Survey on Long Lasting Challenges and New Solutions","version":1},"reference_index":187,"source":"pdf_text","source_observed_at":"2026-05-23T23:47:28.874336Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2406.10861"},"observation_digest":"sha256:1ee5a6870516eff039bb4fc9b47cdafcc7a522eb27672441e19e233b34785aa7","observation_id":"e516d988-8ca5-4b0e-9255-920b1ccc582c","resolution":{"observed_at":"2026-05-23T23:48:39.192081Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-12T14:45:30.276038Z","title":"idlg: Improved deep leakage from gradients","venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2411.14937","last_updated":"2025-08-02T08:52:34Z","snapshot_observed_at":"2026-08-18T08:58:36.783355Z","submitted_at":"2024-11-22T13:49:56Z","title":"Geminio: Language-Guided Gradient Inversion Attacks in Federated Learning","version":2},"reference_index":45,"source":"pdf_text","source_observed_at":"2026-08-12T14:45:30.276038Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2411.14937"},"observation_digest":"sha256:52e0cd51286e97dbd27338072c1d24faab1f464ac2dc41f5be5f2bbfc5040f18","observation_id":"d4d6a573-0931-49a7-b31c-4f6c06093af6","resolution":{"observed_at":"2026-08-12T14:45:30.276038Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-12T05:36:10.273141Z","title":"idlg: Improved deep leakage from gradients","venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2412.00334","last_updated":"2025-08-31T12:40:53Z","snapshot_observed_at":"2026-08-17T23:20:02.902544Z","submitted_at":"2024-11-30T03:20:14Z","title":"EFTViT: Efficient Federated Training of Vision Transformers with Masked Images on Resource-Constrained Clients","version":2},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-08-12T05:36:10.273141Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2412.00334"},"observation_digest":"sha256:19b70b4b26ff212cf920c3127a332ca7bec1542793b7fedafc850ba477838969","observation_id":"3858fcdf-b49f-4b38-bf5b-9625092bab90","resolution":{"observed_at":"2026-08-12T05:36:10.273141Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-11T20:04:51.556545Z","title":"iDLG: Improved deep leakage from gradients,","venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2412.06120","last_updated":"2025-01-11T08:39:49Z","snapshot_observed_at":"2026-08-18T09:00:39.190648Z","submitted_at":"2024-12-09T00:54:00Z","title":"Lightweight Federated Learning with Differential Privacy and Straggler Resilience","version":2},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-11T20:04:51.556545Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2412.06120"},"observation_digest":"sha256:917ad7661fa60a50291931dea191ab0a0ebeda86d207cc9c5d8df998d14ddf95","observation_id":"c30ea56c-d520-4736-9b21-946612ad662c","resolution":{"observed_at":"2026-08-11T20:04:51.556545Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-11T20:00:19.403628Z","title":null,"venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2412.06157","last_updated":"2024-12-09T02:39:58Z","snapshot_observed_at":"2026-08-15T22:22:00.400827Z","submitted_at":"2024-12-09T02:39:58Z","title":"Membership Inference Attacks and Defenses in Federated Learning: A Survey","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-11T20:00:19.403628Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2412.06157"},"observation_digest":"sha256:0c606495698f08e433dd72c3bbbd7303e3d54aedf47688f293f836a950f70170","observation_id":"a0c2cff6-448d-4d14-ab18-229c9139c140","resolution":{"observed_at":"2026-08-11T20:00:19.403628Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-11T19:08:54.628478Z","title":"1 m mX i=1 Li xt+1 i # − L∗ ≤(1 − ηminµ)t+1D + ηmin LM G2 4µ . (32) If we set ηmin ≤ µϵ LM G2 , after O( 1 ϵ log( 1 ϵ )) steps, we have E","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2412.07187","last_updated":"2024-12-10T04:53:42Z","snapshot_observed_at":"2026-08-17T11:01:59.164174Z","submitted_at":"2024-12-10T04:53:42Z","title":"A New Federated Learning Framework Against Gradient Inversion Attacks","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-11T19:08:54.628478Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2412.07187"},"observation_digest":"sha256:7e90df2928896cb397ffea012a2a8d30356a41b93f0b4ef657af88f652efcab1","observation_id":"e458707d-53ff-48b0-b24a-86274ade0241","resolution":{"observed_at":"2026-08-11T19:08:54.628478Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-11T17:48:55.044279Z","title":"R., and Bilen H","venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2412.08544","last_updated":"2024-12-11T17:00:29Z","snapshot_observed_at":"2026-08-21T13:03:27.496779Z","submitted_at":"2024-12-11T17:00:29Z","title":"Training Data Reconstruction: Privacy due to Uncertainty?","version":1},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-11T17:48:55.044279Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2412.08544"},"observation_digest":"sha256:2195c8f9f271ace3b5362ae9fd62ef5031acc28c6b35e2b9dfac9c207f162037","observation_id":"9ba668d7-faab-4215-917d-ef2d0ef46736","resolution":{"observed_at":"2026-08-11T17:48:55.044279Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-11T13:59:14.437789Z","title":"idlg: Improved deep leakage from gradients,","venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2412.12640","last_updated":"2026-07-03T06:57:36Z","snapshot_observed_at":"2026-08-18T15:24:35.415325Z","submitted_at":"2024-12-17T08:03:38Z","title":"GDBR: Label Recovery Attack Against Partial Gradient Encryption in Federated Learning","version":2},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-11T13:59:14.437789Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2412.12640"},"observation_digest":"sha256:e4414c9c729e2c7c0a495a4e0ac8487409f89f284eb8ed212340f5b28c509425","observation_id":"c7eb7356-46d7-4644-9ed4-8e173f384559","resolution":{"observed_at":"2026-08-11T13:59:14.437789Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-11T12:49:37.527741Z","title":"idlg: Improved deep leakage from gradients,","venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2412.13818","last_updated":"2024-12-18T13:05:55Z","snapshot_observed_at":"2026-08-15T18:20:22.313652Z","submitted_at":"2024-12-18T13:05:55Z","title":"Fed-AugMix: Balancing Privacy and Utility via Data Augmentation","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-11T12:49:37.527741Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2412.13818"},"observation_digest":"sha256:6d3f883083af0b83e70e7fbef671cee7ea94293cbd790217f3bdd43a8e8f780a","observation_id":"677ddfc3-fda7-4c12-b958-e11272e7dcda","resolution":{"observed_at":"2026-08-11T12:49:37.527741Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-11T04:45:50.873901Z","title":"idlg: Improved deep leakage from gradients,","venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2412.18513","last_updated":"2024-12-24T15:52:21Z","snapshot_observed_at":"2026-08-17T00:50:34.433111Z","submitted_at":"2024-12-24T15:52:21Z","title":"FedGIG: Graph Inversion from Gradient in Federated Learning","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-11T04:45:50.873901Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2412.18513"},"observation_digest":"sha256:2100341eab393f210fcce375a059979e5696c68ba8e90dd853f91151f0ac182e","observation_id":"582ffb94-32d2-48bb-a54d-9f7cfe1c0f36","resolution":{"observed_at":"2026-08-11T04:45:50.873901Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-10T22:44:11.475973Z","title":null,"venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2501.00842","last_updated":"2025-03-26T15:35:20Z","snapshot_observed_at":"2026-08-18T15:23:29.045992Z","submitted_at":"2025-01-01T13:46:11Z","title":"A Survey of Secure Semantic Communications","version":2},"reference_index":96,"source":"pdf_text","source_observed_at":"2026-08-10T22:44:11.475973Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2501.00842"},"observation_digest":"sha256:d75461ac9a22af55e4aebdfcc4efe252ee59c5e4c2caf6261a31b6a406b02118","observation_id":"6635b7c8-5575-4236-a654-f74e57eadfe7","resolution":{"observed_at":"2026-08-10T22:44:11.475973Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-10T15:06:09.184625Z","title":"R., and Bilen, H","venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2501.14652","last_updated":"2025-06-27T17:22:45Z","snapshot_observed_at":"2026-08-15T12:58:52.064127Z","submitted_at":"2025-01-24T17:18:46Z","title":"Decoupled SGDA for Games with Intermittent Strategy Communication","version":2},"reference_index":62,"source":"arxiv_source","source_observed_at":"2026-08-10T15:06:09.184625Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2501.14652"},"observation_digest":"sha256:e9bb1cc32c655f0887112b9e9b98fd430b9bf520408bd30507ad7dce9b5265b2","observation_id":"8a484179-2747-4884-b7cf-2f808e1d4cfd","resolution":{"observed_at":"2026-08-10T15:06:09.184625Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-10T14:05:45.463438Z","title":"idlg: Improved deep leakage from gradients,","venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2501.15718","last_updated":"2025-01-27T01:06:23Z","snapshot_observed_at":"2026-08-17T23:11:53.730745Z","submitted_at":"2025-01-27T01:06:23Z","title":"CENSOR: Defense Against Gradient Inversion via Orthogonal Subspace Bayesian Sampling","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-10T14:05:45.463438Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2501.15718"},"observation_digest":"sha256:70646d434c5912abd5d4854d1ed64a89b14280bb41d2a43476bd1c4cca87c47f","observation_id":"6449b9c6-659e-4ff6-a9da-a5e9234fdb9e","resolution":{"observed_at":"2026-08-10T14:05:45.463438Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-09T21:58:42.705162Z","title":", Mopuri , K.R","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2501.18934","last_updated":"2025-04-30T06:30:46Z","snapshot_observed_at":"2026-08-19T19:43:02.589534Z","submitted_at":"2025-01-31T07:32:12Z","title":"Deep Learning Model Inversion Attacks and Defenses: A Comprehensive Survey","version":2},"reference_index":187,"source":"arxiv_source","source_observed_at":"2026-08-09T21:58:42.705162Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2501.18934"},"observation_digest":"sha256:bb0fb4309cddbef326b4dca4ca510d36cd0ea33816dec946ed180c2fddd5704b","observation_id":"dea2cfe7-f247-4b09-a536-660e46a09e57","resolution":{"observed_at":"2026-08-09T21:58:42.705162Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-09T10:44:58.547558Z","title":"G., Zhao, C., Sun, Q., Tang, Y ., Qian, F., and Kurths, J","venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2502.02913","last_updated":"2025-02-12T04:59:16Z","snapshot_observed_at":"2026-08-15T01:59:58.010223Z","submitted_at":"2025-02-05T06:20:20Z","title":"Real-Time Privacy Risk Measurement with Privacy Tokens for Gradient Leakage","version":4},"reference_index":2024,"source":"pdf_text","source_observed_at":"2026-08-09T10:44:58.547558Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2502.02913"},"observation_digest":"sha256:cb07c7eb5fbe625974451dfc6d67cf300af48a2f61e2bba122f19f5c21e41de3","observation_id":"9259e078-b24d-4a83-8f2a-2fb6509493fd","resolution":{"observed_at":"2026-08-09T10:44:58.547558Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-09T06:02:19.213773Z","title":"idlg: Improved deep leakage from gradients,","venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2502.03092","last_updated":"2025-02-05T11:31:21Z","snapshot_observed_at":"2026-08-15T21:29:22.906415Z","submitted_at":"2025-02-05T11:31:21Z","title":"E-3SFC: Communication-Efficient Federated Learning with Double-way Features Synthesizing","version":1},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-08-09T06:02:19.213773Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2502.03092"},"observation_digest":"sha256:15510f4672e5c581808e6e9fe2c2563aad627060822cc6bfe817cff2acb834df","observation_id":"367010bd-da7f-44ca-bf99-55597acc56c7","resolution":{"observed_at":"2026-08-09T06:02:19.213773Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-08T23:53:24.221834Z","title":"idlg: Improved deep leakage from gradients,","venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2502.04045","last_updated":"2025-02-06T13:04:25Z","snapshot_observed_at":"2026-08-21T07:33:08.159913Z","submitted_at":"2025-02-06T13:04:25Z","title":"Comparing privacy notions for protection against reconstruction attacks in machine learning","version":1},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-08T23:53:24.221834Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2502.04045"},"observation_digest":"sha256:b4cb8244c257eed3fb30a0383b52b8d7edb6740adece977715c618417140d047","observation_id":"5e7b6d6c-2e8b-444c-8839-948424f204b4","resolution":{"observed_at":"2026-08-08T23:53:24.221834Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-08T23:33:38.862873Z","title":"idlg: Improved deep leakage from gradients,","venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2502.04106","last_updated":"2025-04-10T02:55:11Z","snapshot_observed_at":"2026-08-17T15:39:25.831737Z","submitted_at":"2025-02-06T14:31:14Z","title":"The Gradient Puppeteer: Adversarial Domination in Gradient Leakage Attacks through Model Poisoning","version":2},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-08T23:33:38.862873Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2502.04106"},"observation_digest":"sha256:468dc98cfe1080ad18b8511c1e82afe36ed41f2628c1f5f2f09e0f1d049abf25","observation_id":"8ebbd8c0-ccaf-4a72-85b1-baf02330bff0","resolution":{"observed_at":"2026-08-08T23:33:38.862873Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-15T23:22:16.551840Z","title":null,"venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2505.04889","last_updated":"2025-05-08T01:50:27Z","snapshot_observed_at":"2026-08-21T09:09:29.570722Z","submitted_at":"2025-05-08T01:50:27Z","title":"FedRE: Robust and Effective Federated Learning with Privacy Preference","version":1},"reference_index":45,"source":"pdf_text","source_observed_at":"2026-08-15T23:22:16.551840Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2505.04889"},"observation_digest":"sha256:4024169f04f929c12b6d9e419620f0315a892d1fdf49a9e644865fe9d1c309cb","observation_id":"8bf01601-721e-4bda-879e-3f1eea451019","resolution":{"observed_at":"2026-08-15T23:22:16.551840Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-15T21:20:18.269198Z","title":"idlg: Improved deep leakage from gradients, 2020","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2505.10264","last_updated":"2025-09-05T11:54:06Z","snapshot_observed_at":"2026-08-18T09:03:03.086876Z","submitted_at":"2025-05-15T13:16:32Z","title":"Cutting Through Privacy: A Hyperplane-Based Data Reconstruction Attack in Federated Learning","version":2},"reference_index":37,"source":"arxiv_source","source_observed_at":"2026-08-15T21:20:18.269198Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2505.10264"},"observation_digest":"sha256:83a0e275ab45fee30d2d206a4c3f423ed4cb631e3003bb430842412d7731fd82","observation_id":"5f8c32ee-06d8-43af-b4d1-5c4074b33c76","resolution":{"observed_at":"2026-08-15T21:20:18.269198Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-07T15:34:38.508551Z","title":"iDLG: Improved deep leakage from gradients,","venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2505.14797","last_updated":"2025-05-20T18:08:15Z","snapshot_observed_at":"2026-08-17T11:56:20.318280Z","submitted_at":"2025-05-20T18:08:15Z","title":"Efficient Privacy-Preserving Cross-Silo Federated Learning with Multi-Key Homomorphic Encryption","version":1},"reference_index":48,"source":"pdf_text","source_observed_at":"2026-08-07T15:34:38.508551Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2505.14797"},"observation_digest":"sha256:34c1a0e352c4e4f022023fd76b880558792af877d32065cfc151d24b68cda918","observation_id":"4a081491-2b1c-4d4d-9de4-4aeab883776e","resolution":{"observed_at":"2026-08-07T15:34:38.508551Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-07T14:13:34.161138Z","title":"idlg: Improved deep leakage from gradients,","venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2505.19823","last_updated":"2025-05-26T11:00:31Z","snapshot_observed_at":"2026-08-16T10:58:26.355803Z","submitted_at":"2025-05-26T11:00:31Z","title":"LAPA-based Dynamic Privacy Optimization for Wireless Federated Learning in Heterogeneous Environments","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-07T14:13:34.161138Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2505.19823"},"observation_digest":"sha256:d1c974be1acd9a4163f1f685a18d8881b2052d293ec12b3db9adee275030d641","observation_id":"8811c431-f7e2-4060-bf9c-a28a1f764a86","resolution":{"observed_at":"2026-08-07T14:13:34.161138Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-07T14:06:27.811626Z","title":"idlg: Improved deep leakage from gradients","venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2505.20026","last_updated":"2025-05-26T14:17:00Z","snapshot_observed_at":"2026-08-18T08:58:38.250560Z","submitted_at":"2025-05-26T14:17:00Z","title":"Gradient Inversion Transcript: Leveraging Robust Generative Priors to Reconstruct Training Data from Gradient Leakage","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-07T14:06:27.811626Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2505.20026"},"observation_digest":"sha256:2fdf66a721190e2b2f9a4e286d96b842ce3fd5d16cb36b1ee14a64208214171f","observation_id":"c339bb25-a87d-48e9-9333-c7eea890e729","resolution":{"observed_at":"2026-08-07T14:06:27.811626Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-07T13:48:11.352979Z","title":null,"venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2505.20924","last_updated":"2025-08-07T07:56:57Z","snapshot_observed_at":"2026-08-15T15:00:23.548926Z","submitted_at":"2025-05-27T09:13:40Z","title":"Label Leakage in Federated Inertial-based Human Activity Recognition","version":2},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-07T13:48:11.352979Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2505.20924"},"observation_digest":"sha256:b1bf42f533f75a82b5f3e8da67a7570733b87d5d7bcf66813da2c8c83d77b0dc","observation_id":"eef6faaf-4576-429d-a835-9479a35b850f","resolution":{"observed_at":"2026-08-07T13:48:11.352979Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-07T13:14:09.623844Z","title":null,"venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2505.22447","last_updated":"2025-05-28T15:09:56Z","snapshot_observed_at":"2026-08-22T19:33:50.420999Z","submitted_at":"2025-05-28T15:09:56Z","title":"Privacy-preserving Prompt Personalization in Federated Learning for Multimodal Large Language Models","version":1},"reference_index":67,"source":"pdf_text","source_observed_at":"2026-08-07T13:14:09.623844Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2505.22447"},"observation_digest":"sha256:c0157d259af7cd7b5d55187f4efe33446e5c223a1c849bc96ce73b115bd47da5","observation_id":"f52c5cfd-d0a2-4684-aad2-ec3a2f3a5dd0","resolution":{"observed_at":"2026-08-07T13:14:09.623844Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-07T11:41:15.909897Z","title":"idlg: Improved deep leakage from gradients","venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2506.01777","last_updated":"2025-06-02T15:20:54Z","snapshot_observed_at":"2026-08-13T08:45:30.178716Z","submitted_at":"2025-06-02T15:20:54Z","title":"DRAUN: An Algorithm-Agnostic Data Reconstruction Attack on Federated Unlearning Systems","version":1},"reference_index":57,"source":"pdf_text","source_observed_at":"2026-08-07T11:41:15.909897Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2506.01777"},"observation_digest":"sha256:7f6956765d0c19869bf34f3b34600100c8fa7041bd600b7769bb6d5d1e2b1886","observation_id":"1bce2871-dfe9-49a9-b72d-4540451298e8","resolution":{"observed_at":"2026-08-07T11:41:15.909897Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-07T05:26:50.125905Z","title":"idlg: Improved deep leakage from gradients,","venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2506.07894","last_updated":"2025-06-09T16:12:18Z","snapshot_observed_at":"2026-08-22T06:35:43.192199Z","submitted_at":"2025-06-09T16:12:18Z","title":"Secure Distributed Learning for CAVs: Defending Against Gradient Leakage with Leveled Homomorphic Encryption","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-07T05:26:50.125905Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2506.07894"},"observation_digest":"sha256:773341c3f0a83dd88704b0fba709cad0f88c7232bda7fe1470e0983bbed03b55","observation_id":"346203f9-2537-48b2-80d9-08013fda0c07","resolution":{"observed_at":"2026-08-07T05:26:50.125905Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-07T05:21:07.883060Z","title":"idlg: Improved deep leakage from gradients.arXiv preprint arXiv:2001.02610, 2020","venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2506.08435","last_updated":"2025-06-10T04:17:06Z","snapshot_observed_at":"2026-08-20T06:50:04.600386Z","submitted_at":"2025-06-10T04:17:06Z","title":"Boosting Gradient Leakage Attacks: Data Reconstruction in Realistic FL Settings","version":1},"reference_index":56,"source":"pdf_text","source_observed_at":"2026-08-07T05:21:07.883060Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2506.08435"},"observation_digest":"sha256:5623a103f691e024544086528292e2359d14ead5b18273cf8ddc83cf45f4cc02","observation_id":"5d98bd3f-28a8-4af1-a2fa-5b5867e4f17f","resolution":{"observed_at":"2026-08-07T05:21:07.883060Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-07T04:17:46.220200Z","title":"iDLG: Im- proved deep leakage from gradients","venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2506.11413","last_updated":"2025-06-13T02:23:41Z","snapshot_observed_at":"2026-08-14T21:22:31.316616Z","submitted_at":"2025-06-13T02:23:41Z","title":"Byzantine Outside, Curious Inside: Reconstructing Data Through Malicious Updates","version":1},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-07T04:17:46.220200Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2506.11413"},"observation_digest":"sha256:c0b963a20fbe14d685fa3f84ca73bfaac9e00f37f829c184fba26f0cb7870935","observation_id":"3353c9c9-5495-4978-a5f3-d677f1cea5a3","resolution":{"observed_at":"2026-08-07T04:17:46.220200Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-15T19:51:57.363769Z","title":"idlg: Improved deep leakage from gradients","venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2506.15181","last_updated":"2025-06-18T06:53:52Z","snapshot_observed_at":"2026-08-18T08:58:50.613750Z","submitted_at":"2025-06-18T06:53:52Z","title":"ImprovDML: Improved Trade-off in Private Byzantine-Resilient Distributed Machine Learning","version":1},"reference_index":53,"source":"pdf_text","source_observed_at":"2026-08-15T19:51:57.363769Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2506.15181"},"observation_digest":"sha256:d6016c9f77e1b4e4a19c190254a3ffd1b7841c8b61d368d82f81674d7a21a2dd","observation_id":"9a2ba5da-c2ec-4b9e-b4e6-12faee81ae17","resolution":{"observed_at":"2026-08-15T19:51:57.363769Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-07T12:21:30.616801Z","title":", author Mopuri, K.R","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2506.15711","last_updated":"2025-05-30T17:58:57Z","snapshot_observed_at":"2026-08-16T10:58:25.679305Z","submitted_at":"2025-05-30T17:58:57Z","title":"Shadow defense against gradient inversion attack in federated learning","version":1},"reference_index":40,"source":"arxiv_source","source_observed_at":"2026-08-07T12:21:30.616801Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2506.15711"},"observation_digest":"sha256:c70993a9242f140af86e7656b8ec3e1597e9d78ec44bdfb4e675a90101b65237","observation_id":"fe7f97c7-75a7-449c-9723-6370019b950f","resolution":{"observed_at":"2026-08-07T12:21:30.616801Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-15T18:46:03.203888Z","title":"idlg: Improved deep leakage from gradients,","venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2506.19260","last_updated":"2026-08-19T13:46:55Z","snapshot_observed_at":"2026-08-22T23:11:31.857080Z","submitted_at":"2025-06-24T02:42:08Z","title":"Topology-Aware Differential Privacy in Hierarchical Federated Learning","version":2},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-15T18:46:03.203888Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2506.19260"},"observation_digest":"sha256:a34e392485ee752756bb93dbeccea0d85e4ae1ba45abfc700d0fa2b59e25e413","observation_id":"3c2b4421-646b-4a62-b9a1-e685250f09c9","resolution":{"observed_at":"2026-08-15T18:46:03.203888Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-06T22:46:42.081034Z","title":"arXiv preprint arXiv:2001.02610 (2020) 20 Fei Wang and Baochun Li","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2506.20651","last_updated":"2025-06-25T17:49:26Z","snapshot_observed_at":"2026-08-15T22:34:56.105573Z","submitted_at":"2025-06-25T17:49:26Z","title":"Hear No Evil: Detecting Gradient Leakage by Malicious Servers in Federated Learning","version":1},"reference_index":48,"source":"pdf_text","source_observed_at":"2026-08-06T22:46:42.081034Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2506.20651"},"observation_digest":"sha256:673fd519bf5cde9136af3e07809cce777b54d9b5f870cd69b92fd267d94316b0","observation_id":"7cf528ff-6330-4622-8d9f-48701264221b","resolution":{"observed_at":"2026-08-06T22:46:42.081034Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-06T17:56:02.131700Z","title":"idlg: Improved deep leakage from gradients,","venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2507.09602","last_updated":"2025-07-13T12:16:43Z","snapshot_observed_at":"2026-08-17T10:59:40.332722Z","submitted_at":"2025-07-13T12:16:43Z","title":"DRAGD: A Federated Unlearning Data Reconstruction Attack Based on Gradient Differences","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-06T17:56:02.131700Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2507.09602"},"observation_digest":"sha256:20ce611781baab4292d61b001d454c1b04ae1e5c22e421b87170a8729ce97920","observation_id":"793c06aa-fdab-4f55-a089-40ecff933035","resolution":{"observed_at":"2026-08-06T17:56:02.131700Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-06T13:56:22.585335Z","title":null,"venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2507.19964","last_updated":"2025-07-26T14:32:38Z","snapshot_observed_at":"2026-08-18T08:58:37.694152Z","submitted_at":"2025-07-26T14:32:38Z","title":"Who Owns This Sample: Cross-Client Membership Inference Attack in Federated Graph Neural Networks","version":1},"reference_index":59,"source":"pdf_text","source_observed_at":"2026-08-06T13:56:22.585335Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2507.19964"},"observation_digest":"sha256:14969f56487a6dbadaa9e675f308de1c15e40c89b7db94d5616c7cb11ae83c2d","observation_id":"b48b962c-03c8-4aee-8e83-9bc5f5db3f29","resolution":{"observed_at":"2026-08-06T13:56:22.585335Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-15T17:45:36.612674Z","title":null,"venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2507.21198","last_updated":"2025-07-28T06:06:29Z","snapshot_observed_at":"2026-08-21T09:09:26.301354Z","submitted_at":"2025-07-28T06:06:29Z","title":"Uncovering Gradient Inversion Risks in Practical Language Model Training","version":1},"reference_index":54,"source":"pdf_text","source_observed_at":"2026-08-15T17:45:36.612674Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2507.21198"},"observation_digest":"sha256:a638461354fd6a584c547af24c944d4365914d983646ae16cc099d66bf3c1a7e","observation_id":"701a94a3-0701-4005-a4fa-74193cdeacf1","resolution":{"observed_at":"2026-08-15T17:45:36.612674Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-06T11:55:31.884037Z","title":"idlg: Improved deep leakage from gradients","venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2507.22330","last_updated":"2025-07-30T02:24:26Z","snapshot_observed_at":"2026-08-21T21:52:15.726264Z","submitted_at":"2025-07-30T02:24:26Z","title":"Hypernetworks for Model-Heterogeneous Personalized Federated Learning","version":1},"reference_index":82,"source":"pdf_text","source_observed_at":"2026-08-06T11:55:31.884037Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2507.22330"},"observation_digest":"sha256:102f4b021ae239b84af2ab35f207e05647d57cd1fd4521d3a32f342b88aac167","observation_id":"8b6b1013-0c0b-4f4f-8733-ef2009a49fbb","resolution":{"observed_at":"2026-08-06T11:55:31.884037Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-06T10:58:06.094314Z","title":"R.; and Bilen, H","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2507.23291","last_updated":"2025-08-03T23:23:03Z","snapshot_observed_at":"2026-08-15T20:46:11.551277Z","submitted_at":"2025-07-31T07:09:52Z","title":"Evaluating the Dynamics of Membership Privacy in Deep Learning","version":2},"reference_index":51,"source":"arxiv_source","source_observed_at":"2026-08-06T10:58:06.094314Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2507.23291"},"observation_digest":"sha256:6950f1e6ad8c41d63bb2a7d84efedca9e60af780db90e19acf97fcc0be56b114","observation_id":"cac595b2-1d42-4488-8aa0-59d57798c100","resolution":{"observed_at":"2026-08-06T10:58:06.094314Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-05T22:38:58.747371Z","title":"arXiv preprint arXiv:2001.02610 (2020)","venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2508.06789","last_updated":"2025-08-09T02:38:24Z","snapshot_observed_at":"2026-08-18T08:59:19.509957Z","submitted_at":"2025-08-09T02:38:24Z","title":"Label Inference Attacks against Federated Unlearning","version":1},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-05T22:38:58.747371Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2508.06789"},"observation_digest":"sha256:0b09116985767341f1619d5254faa3d7acdf5ec4f230a27f4c6a3335b7be0dc7","observation_id":"faf0b1c4-2330-4172-9e44-d54049a5677f","resolution":{"observed_at":"2026-08-05T22:38:58.747371Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-05T18:12:38.412145Z","title":null,"venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2508.15031","last_updated":"2025-08-27T05:10:26Z","snapshot_observed_at":"2026-08-14T23:52:31.572833Z","submitted_at":"2025-08-20T19:49:59Z","title":"A Systematic Survey of Model Extraction Attacks and Defenses: State-of-the-Art and Perspectives","version":2},"reference_index":267,"source":"pdf_text","source_observed_at":"2026-08-05T18:12:38.412145Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2508.15031"},"observation_digest":"sha256:269ca8da60edaf33843516fa68624595e5641bbccc5f4d2b6cf53d95db8fb225","observation_id":"e6b81c76-7d15-4198-abdf-1bc4dfe94fe5","resolution":{"observed_at":"2026-08-05T18:12:38.412145Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-04T21:12:52.224737Z","title":"IDLG: Improved deep leakage from gradients","venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2509.08195","last_updated":"2025-09-09T23:59:20Z","snapshot_observed_at":"2026-08-18T23:02:38.802679Z","submitted_at":"2025-09-09T23:59:20Z","title":"Sketched Gaussian Mechanism for Private Federated Learning","version":1},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-04T21:12:52.224737Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2509.08195"},"observation_digest":"sha256:77565bdcfb5a514577fa8d3cd9504261356b8079bcdceedc62a2bad134bef006","observation_id":"185ded53-f506-4cd6-b3f4-6b941656c205","resolution":{"observed_at":"2026-08-04T21:12:52.224737Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-04T19:11:57.447593Z","title":"idlg: Improved deep leakage from gradients","venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2509.09485","last_updated":"2025-09-12T01:27:15Z","snapshot_observed_at":"2026-08-20T01:40:18.656100Z","submitted_at":"2025-09-11T14:17:04Z","title":"Balancing Utility and Privacy: Dynamically Private SGD with Random Projection","version":2},"reference_index":53,"source":"arxiv_source","source_observed_at":"2026-08-04T19:11:57.447593Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2509.09485"},"observation_digest":"sha256:1ac14bd9d3079481bcbe1522f167e852981c38cfff13fc40f80719111706eaad","observation_id":"b9b00dbc-8db5-447d-83b0-86992fbdb8e6","resolution":{"observed_at":"2026-08-04T19:11:57.447593Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-04T19:43:59.954962Z","title":"idlg: Improved deep leakage from gradients.arXiv preprint arXiv:2001.02610, 2020","venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2509.09742","last_updated":"2025-09-11T04:34:42Z","snapshot_observed_at":"2026-08-15T10:48:02.610448Z","submitted_at":"2025-09-11T04:34:42Z","title":"Images in Motion?: A First Look into Video Leakage in Collaborative Deep Learning","version":1},"reference_index":44,"source":"pdf_text","source_observed_at":"2026-08-04T19:43:59.954962Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2509.09742"},"observation_digest":"sha256:e76db8248692b1b91b3820106902bbfca4467a9eedf7c9b7aeabdf73fbae79fe","observation_id":"e4cd18d0-5e29-4f00-80fc-fe3627baa527","resolution":{"observed_at":"2026-08-04T19:43:59.954962Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":"2001.02610","doi":"10.48550/arxiv.2001.02610","metadata_source":"pith","pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"iDLG: Improved deep leakage from gradients","venue":"cs.LG","work_id":"611e14b3-8402-48e3-92f8-d9356cbdc89d","year":2020},"citing_paper":{"arxiv_id":"2604.06297","last_updated":"2026-04-07T17:19:54Z","snapshot_observed_at":"2026-08-13T01:18:08.432071Z","submitted_at":"2026-04-07T17:19:54Z","title":"FedSpy-LLM: Towards Scalable and Generalizable Data Reconstruction Attacks from Gradients on LLMs","version":1},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-05-10T19:04:41.807582Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2604.06297"},"observation_digest":"sha256:5670e87636fd81bfa2467ecbfd8cdcfdcc95b009ac49e9b6144d5d77069ee7e4","observation_id":"4c3968b7-585c-419b-a899-7f3b2993ed2e","resolution":{"observed_at":"2026-05-10T23:30:51.355928Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":"2001.02610","doi":"10.48550/arxiv.2001.02610","metadata_source":"pith","pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"iDLG: Improved deep leakage from gradients","venue":"cs.LG","work_id":"611e14b3-8402-48e3-92f8-d9356cbdc89d","year":2020},"citing_paper":{"arxiv_id":"2604.07125","last_updated":"2026-06-06T12:46:41Z","snapshot_observed_at":"2026-08-16T14:52:04.639272Z","submitted_at":"2026-04-08T14:19:39Z","title":"Scalable and Private Federated Learning Using Distributed Differential Privacy and Secure Aggregation","version":1},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-05-10T17:54:31.930947Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2604.07125"},"observation_digest":"sha256:8832ed9c7c5de42c0074161654ee6d3b0dcc802094b0ca41b43028f5ad92aed0","observation_id":"c3226fbf-8237-45ef-bfd7-1b73afd9a571","resolution":{"observed_at":"2026-05-11T05:51:11.037065Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":"2001.02610","doi":"10.48550/arxiv.2001.02610","metadata_source":"pith","pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"iDLG: Improved deep leakage from gradients","venue":"cs.LG","work_id":"611e14b3-8402-48e3-92f8-d9356cbdc89d","year":2020},"citing_paper":{"arxiv_id":"2604.14833","last_updated":"2026-04-16T10:10:16Z","snapshot_observed_at":"2026-08-14T04:57:06.882194Z","submitted_at":"2026-04-16T10:10:16Z","title":"Federated User Behavior Modeling for Privacy-Preserving LLM Recommendation","version":1},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-05-10T10:08:51.386465Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2604.14833"},"observation_digest":"sha256:0d187d5565f787623f12d09446720896976111235df2c7bbbee007750c516904","observation_id":"385fe5d8-4359-4a33-a158-5cbc609cd490","resolution":{"observed_at":"2026-05-10T10:09:06.489828Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":"2001.02610","doi":"10.48550/arxiv.2001.02610","metadata_source":"pith","pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"iDLG: Improved deep leakage from gradients","venue":"cs.LG","work_id":"611e14b3-8402-48e3-92f8-d9356cbdc89d","year":2020},"citing_paper":{"arxiv_id":"2604.16606","last_updated":"2026-04-17T18:00:58Z","snapshot_observed_at":"2026-07-06T23:03:52.631613Z","submitted_at":"2026-04-17T18:00:58Z","title":"SafeLM: Unified Privacy-Aware Optimization for Trustworthy Federated Large Language Models","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-05-10T08:05:43.387055Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2604.16606"},"observation_digest":"sha256:5728bb64e8017e255800602ebbbb37a3bf6c060c59ff71c81a3b77ac2f6283cc","observation_id":"f9b0f2d4-1ef0-4c67-b67b-9b687ca25efb","resolution":{"observed_at":"2026-05-10T09:13:30.504722Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":"2001.02610","doi":"10.48550/arxiv.2001.02610","metadata_source":"pith","pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"iDLG: Improved deep leakage from gradients","venue":"cs.LG","work_id":"611e14b3-8402-48e3-92f8-d9356cbdc89d","year":2020},"citing_paper":{"arxiv_id":"2605.08288","last_updated":"2026-05-08T08:15:47Z","snapshot_observed_at":"2026-07-06T23:20:33.816373Z","submitted_at":"2026-05-08T08:15:47Z","title":"UMEDA: Unified Multi-modal Efficient Data Fusion for Privacy-Preserving Graph Federated Learning via Spectral-Gated Attention and Diffusion-Based Operator Alignment","version":1},"reference_index":62,"source":"pdf_text","source_observed_at":"2026-05-12T02:47:38.972072Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2605.08288"},"observation_digest":"sha256:6bd3d449f120387c7401bc25499966ecdb0d525f41ff1787b47b6ae509d630c0","observation_id":"efd10b79-1019-473d-b0dd-6964dcada28c","resolution":{"observed_at":"2026-05-12T02:51:18.040297Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":"2001.02610","doi":"10.48550/arxiv.2001.02610","metadata_source":"pith","pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"iDLG: Improved deep leakage from gradients","venue":"cs.LG","work_id":"611e14b3-8402-48e3-92f8-d9356cbdc89d","year":2020},"citing_paper":{"arxiv_id":"2605.12168","last_updated":"2026-05-12T14:16:05Z","snapshot_observed_at":"2026-08-12T13:39:42.912938Z","submitted_at":"2026-05-12T14:16:05Z","title":"On What We Can Learn from Low-Resolution Data","version":1},"reference_index":110,"source":"arxiv_source","source_observed_at":"2026-05-13T05:28:50.993737Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2605.12168"},"observation_digest":"sha256:f8b3fa905629374c9b022e6fa06836a9006c8fea133206a65e4dc6f793c58764","observation_id":"50bb0aff-8a95-4543-b798-49af401145be","resolution":{"observed_at":"2026-05-13T05:32:19.164126Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":"2001.02610","doi":"10.48550/arxiv.2001.02610","metadata_source":"pith","pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"iDLG: Improved deep leakage from gradients","venue":"cs.LG","work_id":"611e14b3-8402-48e3-92f8-d9356cbdc89d","year":2020},"citing_paper":{"arxiv_id":"2605.13265","last_updated":"2026-05-13T09:45:49Z","snapshot_observed_at":"2026-08-16T06:05:36.454130Z","submitted_at":"2026-05-13T09:45:49Z","title":"LightSplit: Practical Privacy-Preserving Split Learning via Orthogonal Projections","version":1},"reference_index":52,"source":"pdf_text","source_observed_at":"2026-05-14T19:46:11.837750Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2605.13265"},"observation_digest":"sha256:b582888bbdf4ea2a734afa089dc982f1507abf6b6a5b0b0b7cf62bb099c763e3","observation_id":"81a4fa43-e5d3-4c2c-baf0-5704d49dc0bc","resolution":{"observed_at":"2026-05-14T19:47:52.990549Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":"2001.02610","doi":"10.48550/arxiv.2001.02610","metadata_source":"pith","pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"iDLG: Improved deep leakage from gradients","venue":"cs.LG","work_id":"611e14b3-8402-48e3-92f8-d9356cbdc89d","year":2020},"citing_paper":{"arxiv_id":"2605.15573","last_updated":"2026-05-15T03:33:20Z","snapshot_observed_at":"2026-08-15T12:14:02.408099Z","submitted_at":"2026-05-15T03:33:20Z","title":"Response-Conditioned Parallel-to-Sequential Orchestration for Multi-Agent Systems","version":1},"reference_index":53,"source":"arxiv_source","source_observed_at":"2026-05-20T19:30:13.469451Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2605.15573"},"observation_digest":"sha256:ec318b1f1bd1a8d01f4dcbf8033ac044b265b46bbe11b9b5c18a6893f76c19d1","observation_id":"0ca3d2a7-390f-4863-b145-bd400d023e18","resolution":{"observed_at":"2026-05-20T19:33:42.207110Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":"2001.02610","doi":"10.48550/arxiv.2001.02610","metadata_source":"pith","pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"iDLG: Improved deep leakage from gradients","venue":"cs.LG","work_id":"611e14b3-8402-48e3-92f8-d9356cbdc89d","year":2020},"citing_paper":{"arxiv_id":"2605.22898","last_updated":"2026-05-21T16:06:00Z","snapshot_observed_at":"2026-08-17T16:31:35.871594Z","submitted_at":"2026-05-21T16:06:00Z","title":"FIRMA: FIbonacci Ring Model Aggregation for Privacy-preserving Federated Learning","version":1},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-05-25T05:32:08.241443Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2605.22898"},"observation_digest":"sha256:aa7ca1b876eadf3e0285abeb64343b04ef275bc5c0da3d8e204fa1edfbd7c82a","observation_id":"60ffd0f9-63f1-4ce7-866e-701576ab1419","resolution":{"observed_at":"2026-05-25T05:36:40.202704Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":"2001.02610","doi":"10.48550/arxiv.2001.02610","metadata_source":"pith","pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"iDLG: Improved deep leakage from gradients","venue":"cs.LG","work_id":"611e14b3-8402-48e3-92f8-d9356cbdc89d","year":2020},"citing_paper":{"arxiv_id":"2605.29845","last_updated":"2026-05-28T12:27:49Z","snapshot_observed_at":"2026-08-18T08:58:34.328611Z","submitted_at":"2026-05-28T12:27:49Z","title":"Local Differential Privacy via Dynamic Quantization in Distributed Online Stochastic Optimization","version":1},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-06-29T05:56:38.098719Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2605.29845"},"observation_digest":"sha256:624ae7e877d73d17948144f6f62ec42f9c2e7b2f7e6383faf2ee249aef2fabaf","observation_id":"10aa6f53-284f-4adf-9993-4592a93cd483","resolution":{"observed_at":"2026-06-29T06:03:08.817359Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":"2001.02610","doi":"10.48550/arxiv.2001.02610","metadata_source":"pith","pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"iDLG: Improved deep leakage from gradients","venue":"cs.LG","work_id":"611e14b3-8402-48e3-92f8-d9356cbdc89d","year":2020},"citing_paper":{"arxiv_id":"2606.00986","last_updated":"2026-05-31T03:54:47Z","snapshot_observed_at":"2026-08-14T16:25:55.646490Z","submitted_at":"2026-05-31T03:54:47Z","title":"Profiling Privacy Preservation Against Gradient Inversion Attacks in Tabular Federated Learning","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-06-28T17:43:10.361804Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2606.00986"},"observation_digest":"sha256:2e9cc5df752018976e447c65eed166e70d45115d89fc0e03922128aa2aaa9a03","observation_id":"2de3c98b-a564-4680-874e-5680ae861a25","resolution":{"observed_at":"2026-07-01T20:46:14.123181Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":"2001.02610","doi":"10.48550/arxiv.2001.02610","metadata_source":"pith","pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"iDLG: Improved deep leakage from gradients","venue":"cs.LG","work_id":"611e14b3-8402-48e3-92f8-d9356cbdc89d","year":2020},"citing_paper":{"arxiv_id":"2606.04399","last_updated":"2026-06-03T03:27:40Z","snapshot_observed_at":"2026-08-02T04:30:22.599452Z","submitted_at":"2026-06-03T03:27:40Z","title":"DPDL: Towards Differential Privacy Preservation in Decentralized Stochastic Learning on Non-IID Data","version":1},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-06-28T07:42:44.280287Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2606.04399"},"observation_digest":"sha256:534c2b62a1803899ca69225b53495aebf9191f7d100d179c66ebd760b0788897","observation_id":"ad1ddc5c-4c16-4e20-9ff5-ed145e987289","resolution":{"observed_at":"2026-07-02T06:06:41.143091Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":"2001.02610","doi":"10.48550/arxiv.2001.02610","metadata_source":"pith","pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"iDLG: Improved deep leakage from gradients","venue":"cs.LG","work_id":"611e14b3-8402-48e3-92f8-d9356cbdc89d","year":2020},"citing_paper":{"arxiv_id":"2606.10780","last_updated":"2026-06-09T12:33:53Z","snapshot_observed_at":"2026-08-18T09:00:05.697193Z","submitted_at":"2026-06-09T12:33:53Z","title":"Secure Aggregation with Top-K Sparsification in Decentralized Federated Learning","version":1},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-06-27T11:41:26.971107Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2606.10780"},"observation_digest":"sha256:61552f46be1c30217a2b3a4b512bc758fa8674f1ccbe5b348e2888f87497e103","observation_id":"cb47a329-1a50-4a90-953c-32bcc9d9afb7","resolution":{"observed_at":"2026-07-03T07:57:44.288500Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":"2001.02610","doi":"10.48550/arxiv.2001.02610","metadata_source":"pith","pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"iDLG: Improved deep leakage from gradients","venue":"cs.LG","work_id":"611e14b3-8402-48e3-92f8-d9356cbdc89d","year":2020},"citing_paper":{"arxiv_id":"2606.17110","last_updated":"2026-06-15T07:04:01Z","snapshot_observed_at":"2026-08-05T23:11:05.452983Z","submitted_at":"2026-06-15T07:04:01Z","title":"Loss Landscape Poisoning: Targeted Extraction of Unseen Training Data from LLMs","version":1},"reference_index":91,"source":"pdf_text","source_observed_at":"2026-06-27T03:59:30.468854Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2606.17110"},"observation_digest":"sha256:9a889652c0a6c7a36b97ff0ea93c5bc680a569b283f7f128c9b9276f4b55ecd0","observation_id":"a397fed1-909a-49ef-b449-80e4100ba73c","resolution":{"observed_at":"2026-07-03T17:38:43.793108Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":"2001.02610","doi":"10.48550/arxiv.2001.02610","metadata_source":"pith","pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"iDLG: Improved deep leakage from gradients","venue":"cs.LG","work_id":"611e14b3-8402-48e3-92f8-d9356cbdc89d","year":2020},"citing_paper":{"arxiv_id":"2606.18312","last_updated":"2026-06-16T10:24:40Z","snapshot_observed_at":"2026-08-15T18:03:06.064443Z","submitted_at":"2026-06-16T10:24:40Z","title":"TIGER: Inverting Transformer Gradients via Embedding-Subspace Distance Optimization","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-06-27T00:25:15.689389Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2606.18312"},"observation_digest":"sha256:9fc30ab934bdd7d2d36361ab1e1879837fff8e0344f5414602882e4899342412","observation_id":"9a143d38-99c9-41b0-8331-44697dd203ff","resolution":{"observed_at":"2026-07-03T21:28:59.420722Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":"2001.02610","doi":"10.48550/arxiv.2001.02610","metadata_source":"pith","pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"iDLG: Improved deep leakage from gradients","venue":"cs.LG","work_id":"611e14b3-8402-48e3-92f8-d9356cbdc89d","year":2020},"citing_paper":{"arxiv_id":"2606.20553","last_updated":"2026-06-18T17:58:25Z","snapshot_observed_at":"2026-08-16T06:58:07.806823Z","submitted_at":"2026-06-18T17:58:25Z","title":"From Efficiency to Leakage -- Privacy Backdoor in Federated Language Model Fine-Tuning","version":1},"reference_index":62,"source":"pdf_text","source_observed_at":"2026-06-26T16:51:07.028013Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2606.20553"},"observation_digest":"sha256:0dc4ae9d325e85f52e6bbba976a3a00c9ac6d83ab0603dd2f45d2c943abacacb","observation_id":"77516e72-8a29-4c82-a1c2-5b34ce6550c1","resolution":{"observed_at":"2026-07-04T04:39:35.034868Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":"2001.02610","doi":"10.48550/arxiv.2001.02610","metadata_source":"pith","pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"iDLG: Improved deep leakage from gradients","venue":"cs.LG","work_id":"611e14b3-8402-48e3-92f8-d9356cbdc89d","year":2020},"citing_paper":{"arxiv_id":"2606.22928","last_updated":"2026-06-22T07:05:53Z","snapshot_observed_at":"2026-08-17T14:13:58.617918Z","submitted_at":"2026-06-22T07:05:53Z","title":"HADES: Privacy-Preserving Federated Learning via Selective Feature Encryption and Hybrid Model Fusion","version":1},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-06-26T08:00:19.005588Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2606.22928"},"observation_digest":"sha256:839012a47ea027d5250ffcda1145f4fa5d4dbe1dc9df104e706cec5d11949a5e","observation_id":"6d5ae6e5-03f0-49cf-9d28-130a677145c3","resolution":{"observed_at":"2026-07-04T11:29:50.211214Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":"2001.02610","doi":"10.48550/arxiv.2001.02610","metadata_source":"pith","pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"iDLG: Improved deep leakage from gradients","venue":"cs.LG","work_id":"611e14b3-8402-48e3-92f8-d9356cbdc89d","year":2020},"citing_paper":{"arxiv_id":"2606.23276","last_updated":"2026-06-24T01:41:10Z","snapshot_observed_at":"2026-08-13T18:49:43.934849Z","submitted_at":"2026-06-22T12:53:54Z","title":"Exposing the Illusion of Erasure in Knowledge Editing for LLMs","version":2},"reference_index":47,"source":"pdf_text","source_observed_at":"2026-06-26T09:10:39.422141Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2606.23276"},"observation_digest":"sha256:b9f3dcd97274fc685617c6ebbaf865c0bfe7c74f076b2707e0e3393451893971","observation_id":"d5f9e80f-a1ac-4a49-9589-37f7fdca070c","resolution":{"observed_at":"2026-07-04T10:09:44.315325Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":"2001.02610","doi":"10.48550/arxiv.2001.02610","metadata_source":"pith","pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"iDLG: Improved deep leakage from gradients","venue":"cs.LG","work_id":"611e14b3-8402-48e3-92f8-d9356cbdc89d","year":2020},"citing_paper":{"arxiv_id":"2606.25627","last_updated":"2026-06-24T09:34:27Z","snapshot_observed_at":"2026-08-21T12:28:32.945810Z","submitted_at":"2026-06-24T09:34:27Z","title":"TL++: Accuracy and Privacy Preserving Traversal Learning for Distributed Intelligent Systems","version":1},"reference_index":67,"source":"pdf_text","source_observed_at":"2026-06-25T20:59:16.064037Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2606.25627"},"observation_digest":"sha256:45ef6fb3d9ab9d18dab601948ba1314a03cb5018407d938ef70c443fc3297bd2","observation_id":"8926ec3c-7c16-4cf8-b2cb-1951cd5d2b34","resolution":{"observed_at":"2026-07-04T19:50:10.991604Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":"2001.02610","doi":"10.48550/arxiv.2001.02610","metadata_source":"pith","pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"iDLG: Improved deep leakage from gradients","venue":"cs.LG","work_id":"611e14b3-8402-48e3-92f8-d9356cbdc89d","year":2020},"citing_paper":{"arxiv_id":"2607.07314","last_updated":"2026-07-08T11:55:00Z","snapshot_observed_at":"2026-08-13T18:04:04.049855Z","submitted_at":"2026-07-08T11:55:00Z","title":"FedCVESA: Taking Away Training Data in Federated Learning via Correlation Value Encoding and Segmented Aggregation","version":1},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-07-09T14:34:41.538077Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2607.07314"},"observation_digest":"sha256:20a2bf564085dc5fe442d8edcb09fe7e5cce7d84ccb1fb654757fbfd70be3cef","observation_id":"5a613271-5da4-4c01-8989-f05ded997bc7","resolution":{"observed_at":"2026-07-09T14:36:17.470325Z","resolver_source":"local_arxiv","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-02T02:54:02.313861Z","title":"iDLG: improved deep leakage from gradients","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2607.14205","last_updated":"2026-07-15T17:57:53Z","snapshot_observed_at":"2026-08-19T07:47:28.735173Z","submitted_at":"2026-07-15T17:57:53Z","title":"Privacy Leakage in Federated Learning in Radiology Reports: A Comparative Evaluation of Tokenizer-Driven Privacy Risks","version":1},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-02T02:54:02.313861Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2607.14205"},"observation_digest":"sha256:4fa9f2ac9b5a4288c23cdd728d8ddbcfb9f4dafd8f0cabbf9d2787c402fd7d9d","observation_id":"bf6ba427-6506-4db5-adca-b951ec2a362f","resolution":{"observed_at":"2026-08-02T02:54:02.313861Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-01T21:49:07.116513Z","title":"idlg: Improved deep leakage from gradients,","venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2607.15970","last_updated":"2026-07-17T14:04:59Z","snapshot_observed_at":"2026-08-17T13:16:51.737295Z","submitted_at":"2026-07-17T14:04:59Z","title":"Code-Poisoning Property Inference Attacks","version":1},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-01T21:49:07.116513Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2607.15970"},"observation_digest":"sha256:7193008dca77158ead8c887dc98167961bf340585f00c7b974f11212632d9d86","observation_id":"0a888b35-fc3b-44f4-9893-9393b8b71ddf","resolution":{"observed_at":"2026-08-01T21:49:07.116513Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-07-31T11:46:52.426745Z","title":"idlg: Improved deep leakage from gradients, 2020","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2607.24556","last_updated":"2026-07-27T15:29:49Z","snapshot_observed_at":"2026-08-14T10:31:52.566666Z","submitted_at":"2026-07-27T15:29:49Z","title":"BettiSplit: Topology-Guided Privacy-Aware Split Learning Against Feature Inversion and Gradient Leakage","version":1},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-07-31T11:46:52.426745Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2607.24556"},"observation_digest":"sha256:975fe49c1fa98f6c09f598ebd3ec3f7e0507844b54db621922c8b4f9d03a849c","observation_id":"64843254-88f2-48b7-813a-5633b11b36d3","resolution":{"observed_at":"2026-07-31T11:46:52.426745Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-07-31T23:40:03.750019Z","title":null,"venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2607.27886","last_updated":"2026-07-31T01:31:02Z","snapshot_observed_at":"2026-08-18T09:02:49.381766Z","submitted_at":"2026-07-30T09:02:02Z","title":"Don't Trust the AI Ecosystem: Analyzing Privacy Leakage in Compromised Open-Source Components","version":1},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-07-31T23:40:03.750019Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2607.27886"},"observation_digest":"sha256:1d7dcc2917ed79b10e9d8f5594278e4ae0fa309d62245edf865c4241e8e99576","observation_id":"e7817562-e24f-4f78-b3d2-1de6a2c33671","resolution":{"observed_at":"2026-07-31T23:40:03.750019Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-03T01:44:07.275956Z","title":null,"venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2607.27886","last_updated":"2026-07-31T01:31:02Z","snapshot_observed_at":"2026-08-18T09:02:49.381766Z","submitted_at":"2026-07-30T09:02:02Z","title":"Don't Trust the AI Ecosystem: Analyzing Privacy Leakage in Compromised Open-Source Components","version":2},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-03T01:44:07.275956Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2607.27886"},"observation_digest":"sha256:298da06d4f71e53e4207e7646c5b18bbb11eb9fbde30deab8aead3aa591f706e","observation_id":"e2005e9a-e3f1-4669-a8a7-0e3de04654c5","resolution":{"observed_at":"2026-08-03T01:44:07.275956Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-07-31T22:47:13.303660Z","title":"R., & Bilen, H","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2607.27940","last_updated":"2026-07-31T04:21:25Z","snapshot_observed_at":"2026-08-21T03:47:22.260527Z","submitted_at":"2026-07-30T09:49:13Z","title":"TriShield: Zero-Utility-Loss Defense Against Privacy Backdoors in Federated Language Model Fine-Tuning via Orthogonal Gradient Projection and Optimizer State Entanglement","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-07-31T22:47:13.303660Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2607.27940"},"observation_digest":"sha256:366ff2bc14bc0a72e01f89ae97ddd4b912550faa4ca43a8f88423ec208b3e046","observation_id":"6f0b3a3f-445c-4ef2-abc9-c8127fb0c319","resolution":{"observed_at":"2026-07-31T22:47:13.303660Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.02610","snapshot_observed_at":"2026-08-03T01:42:09.787047Z","title":"R., & Bilen, H","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2607.27940","last_updated":"2026-07-31T04:21:25Z","snapshot_observed_at":"2026-08-21T03:47:22.260527Z","submitted_at":"2026-07-30T09:49:13Z","title":"TriShield: Zero-Utility-Loss Defense Against Privacy Backdoors in Federated Language Model Fine-Tuning via Orthogonal Gradient Projection and Optimizer State Entanglement","version":2},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-03T01:42:09.787047Z"},"links":{"cited_paper":"/paper/2001.02610","citing_paper":"/paper/2607.27940"},"observation_digest":"sha256:0a18cba364b7861452bcd7f34ba30356a87c28cf4fca2a5e9f17214697454b5a","observation_id":"9bf33032-04ef-4787-816a-35c64bf3ee88","resolution":{"observed_at":"2026-08-03T01:42:09.787047Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"links":{"evidence":"/evidence","html":"/paper/2001.02610/citation-record","integrity":"/paper/2001.02610/integrity","json":"/paper/2001.02610/citation-record.json","paper":"/paper/2001.02610"},"outbound":[],"paper":{"arxiv_id":"2001.02610","last_updated":"2020-01-08T16:45:09Z","latest_version":1,"primary_category":"cs.LG","snapshot_observed_at":"2026-08-20T00:01:01.707115Z","submitted_at":"2020-01-08T16:45:09Z","title":"iDLG: Improved Deep Leakage from Gradients"},"reference_resolution":{"displayed":0,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":0,"verified_exact":0,"verified_fuzzy":0},"total_outbound_references":0},"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":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"thesis":"As of 23 August 2026, this Paper Citation Record lists 0 of 0 outbound references and 71 inbound Pith citation observations for arXiv:2001.02610."}