{"as_of":"2026-08-11T09:54:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:edf7563fc6f88d82d74800487853237902e4e9b5bbab3e84ac23216f949e52d1","coverage":[{"denominator":33,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":33,"source":"paper_references, paper_reference_links","source_observed_at":"2026-05-21T07:06:34.274842Z","state":"measured"},{"denominator":33,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":33,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-11T06:34:44.6726+00:00","state":"measured"},{"denominator":0,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":0,"source":"paper_references, paper_reference_links","source_observed_at":null,"state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"cited_works","source_observed_at":null,"state":"measured"}],"external_citation_measurements":[],"inbound":[],"links":{"evidence":"/evidence","html":"/paper/2605.20521/citation-record","integrity":"/paper/2605.20521/integrity","json":"/paper/2605.20521/citation-record.json","paper":"/paper/2605.20521"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Calibrating noise to sensitivity in private data analysis","venue":null,"work_id":"a739f6f9-175a-426e-9215-cfd3218c3a58","year":2006},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:2c157e7a085aaef204e26c4c578655191767d27e349e2a3f76a22ed2986c43a5","observation_id":"fe3f90e4-eb02-49e4-8fd6-98255ca17107","resolution":{"observed_at":"2026-05-21T07:09:47.165138Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Differential privacy: A survey of results","venue":null,"work_id":"370a8aea-dcbf-4e99-852c-a6bbebcf4f3a","year":2008},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:4bae3c441779b09d183965cfa1ea9dcf088dbbec90f1f25b690fb27052e51f49","observation_id":"7fcbd85f-dcc4-4b39-a348-890e511ac7c2","resolution":{"observed_at":"2026-05-21T07:09:47.159473Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Differentially private empir- ical risk minimization with input perturbation","venue":null,"work_id":"c678f808-8460-43fe-9205-0569e67937f5","year":2017},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:1e0da774bde8d8eb9c502e1f41e2223b75d7f248e56fdb6883464023d1618d96","observation_id":"036a037a-ea06-4578-9953-8f3db89037fa","resolution":{"observed_at":"2026-05-21T07:09:47.161287Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Certified robustness to adversarial examples with differential privacy","venue":null,"work_id":"27b8b518-607c-41a8-aabc-5d50427b9dfd","year":2019},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:4762faa0fca108bbb0ad5836e409c98bd7291b390f1ba5aa62531faf54620f23","observation_id":"b6c41b2c-0cf1-45eb-a399-150419429f6a","resolution":{"observed_at":"2026-05-21T07:09:47.169449Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":"10.24963/ijcai.2019/660","metadata_source":"openalex","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"Heterogeneous gaussian mechanism: Preserving differential privacy in deep learning with provable robustness","venue":null,"work_id":"5f54e24e-5363-452a-b97c-fb175619de9b","year":2019},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:b762df36b49d91609766ad459eef653be712d26ae26742c15fc5ce1cd5ef6ba9","observation_id":"d6086493-11f4-446b-a953-c79ff3af7d69","resolution":{"observed_at":"2026-05-21T07:09:46.162982Z","resolver_source":"doi","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-05-21T17:22:29.86348+00:00","source":"crossref_status_cache"},{"observed_at":"2026-05-21T17:22:29.86348+00:00","source":"openalex_status_cache"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Differential privacy preservation for deep auto-encoders: an application of human behavior prediction","venue":null,"work_id":"29a40ff8-835f-406b-94d6-d46caa68af71","year":2016},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:263ab1a72fc2855f25cca48070ab2be2ee4020755818b3082baed3a47457e473","observation_id":"738513bf-1e38-412f-b9c5-831c575ec2ed","resolution":{"observed_at":"2026-05-21T07:09:47.150444Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Preserving differential privacy in convolutional deep belief networks","venue":null,"work_id":"347f0743-e501-4bcc-a829-f545af6bfcb3","year":2017},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:3f8b6968ae52f6fb3900369992e320958dcc76471562e57f8c344e48e896be62","observation_id":"56585f6b-0587-4280-b0b7-4d19441f0178","resolution":{"observed_at":"2026-05-21T07:09:47.152227Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Adaptive laplace mechanism: Differential privacy preservation in deep learning","venue":null,"work_id":"6a918a32-224b-4b39-8df7-a9010bd8d755","year":2017},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:18e7f793e8743a05e30eaae6f5ccce41fd9018ecb901b61d8d44ac31917f7f94","observation_id":"5311b344-87b9-469a-bedd-a3e9696efb95","resolution":{"observed_at":"2026-05-21T07:09:47.153956Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Towards practical differentially private convex optimization","venue":null,"work_id":"db561eef-6933-4883-8bdb-cabd83021046","year":2019},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:c379b9b2dc50f0100aaecf2f90cc034ded1a8ae30aaa2c2a6660bb7289dcc336","observation_id":"4271d88c-fe1b-4017-a0ca-fee123d50056","resolution":{"observed_at":"2026-05-21T07:09:47.148537Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"6749.297831","doi":"10.1145/2976749.2978313","metadata_source":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"Goodfellow, H","venue":null,"work_id":"518424d2-1085-4f06-9f9f-1c3aa7913ecb","year":2016},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:5ec69a980ab68459674c3b4a720fbcdea8e2250d0e59085d50eb78b8b6a438c9","observation_id":"7f9339b1-662d-4cc6-b147-89d11d54fcd9","resolution":{"observed_at":"2026-05-21T07:09:46.160060Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"9819.322007","doi":"10.1145/3219819.3220074","metadata_source":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"Concentrated differentially private gradient descent with adaptive per-iteration privacy budget","venue":null,"work_id":"2202efdf-4ad6-4163-a7e7-c19fa708f7d3","year":2018},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:e8a09d59de2d0df71d708b4b5300e0114f9336840b5788dbe19e8d195ca262af","observation_id":"88c5ed19-eea1-409e-bca1-010e9c3db4d6","resolution":{"observed_at":"2026-05-21T07:09:46.167514Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Do not let privacy overbill utility: Gradient embedding perturbation for private learning","venue":null,"work_id":"b960d5e5-32a1-4da3-8592-908cabba6a8b","year":2021},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:9c107cf654ede97e59f6bfce24fe68447e78547ea9b3d1203d19da374badeb56","observation_id":"51d4e67e-1fe0-4d56-84c3-a38fcfe76d50","resolution":{"observed_at":"2026-05-21T07:09:47.157683Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Medical imaging deep learning with differential privacy","venue":null,"work_id":"d65303cd-71ca-4ae6-8474-ce646e7fe10c","year":2021},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:a969036b731fe7e150cf3864920c8b954bc5717cbfaba317d0e726e182de3820","observation_id":"fcf0b2be-5dc9-4262-983b-af8eeaca315f","resolution":{"observed_at":"2026-05-21T07:09:47.144696Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"2506.00660","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Differential privacy for deep learning in medicine","venue":null,"work_id":"eb4dab1e-5429-4050-882a-d635d5a5ea6e","year":2025},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:86dcc9e00005cfc32c39048afa94c2da926d746faf06307bf3bda5122cb03f30","observation_id":"bd6f8966-0b75-4c77-8ec0-8b05da552bcf","resolution":{"observed_at":"2026-05-21T07:09:46.329969Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Analysis of application examples of dif- ferential privacy in deep learning","venue":null,"work_id":"958fd659-ebd9-41c6-9a09-4c63ecdfcd31","year":2021},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:fe23f28926be9bf8bc6ed4632b149623be97fbd306f4a9b5edbcfc4a450d696c","observation_id":"ec9ac697-feb0-4b66-8665-a38159d03f23","resolution":{"observed_at":"2026-05-21T07:09:47.142856Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Mechanism design via differential privacy","venue":null,"work_id":"655592da-6af9-47e8-808a-b8b8ea08b42a","year":2007},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:a9aa07219f81b1e0866776d9571ca7bca7325187d2f9e31154552553a2bc58aa","observation_id":"b41b09b4-7ce1-4997-9b31-ae876e93253d","resolution":{"observed_at":"2026-05-21T07:09:47.141028Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"5804.183586","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Data mining with differential privacy","venue":null,"work_id":"fc252c62-0c88-4a95-98b1-33f9170f15d3","year":2010},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:6c390668eaafecb37327d71076a7f4d969aea17a327544280105f4ddbb065150","observation_id":"14cd9299-b38c-4897-ac63-337da6dbc077","resolution":{"observed_at":"2026-05-21T07:09:46.171259Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":"10.1137/1.9781611973105.101","metadata_source":"openalex","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"Kapralov and K","venue":null,"work_id":"693775c6-6083-4f52-9c92-a08f20680c4a","year":2013},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:cc182dbbc80b291dabe450c383a92f33bbc3eac5cf3c9ad86fb13b17e48439c9","observation_id":"f2f545cb-c953-40a1-b03e-65a104a2deae","resolution":{"observed_at":"2026-05-21T07:09:46.155848Z","resolver_source":"doi","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-05-21T17:22:31.595757+00:00","source":"crossref_status_cache"},{"observed_at":"2026-05-21T17:22:31.595757+00:00","source":"openalex_status_cache"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"6187.323620","doi":"10.14778/3236187.3236200","metadata_source":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"Differentially private hierarchical count-of-counts histograms","venue":"Proceedings of the VLDB Endowment","work_id":"19b2685e-c418-4fe5-b325-11779d18c280","year":2018},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:f8dc9b0c0a540de09814716fed765b7a6ea6a6d78e29d4a71c3d9a1e66c22ba8","observation_id":"a97ec3e1-6aab-4c41-8cce-93d43c259f44","resolution":{"observed_at":"2026-05-21T07:09:46.152735Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Differential privacy without sensitivity","venue":null,"work_id":"b4bf66c3-11ef-4c8e-8008-62d91f6534f4","year":2016},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:4957ee2a12b8239f8e3b1134427913c1f3df0ca5661e506731eb46b2a5a8113c","observation_id":"a7e4a62a-9b7b-4c61-841e-49d7e80d7eed","resolution":{"observed_at":"2026-05-21T07:09:47.134580Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Are normalizing flows the key to unlocking the exponential mechanism? a path through the accuracy-privacy ceiling constraining differentially private ml","venue":null,"work_id":"c7918279-43b6-4a81-b597-e45f70dcec0b","year":2024},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:f0b0f5aee8fac775c4183a034e5fed9fcfeb3062502335a5bf4db418effd462e","observation_id":"63b685f3-2dc1-4d7f-ae23-4c219b206b25","resolution":{"observed_at":"2026-05-21T07:09:47.136621Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"The algorithmic foundations of differential privacy","venue":null,"work_id":"0273156d-7f2f-4ad3-80e7-3e9a8908b5b6","year":2014},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:7e41337a4bd6abc122f408c43ba585585c8a971cee835bdcf18bf80ecd7ab97e","observation_id":"9782cf85-6ccc-4af2-aae2-1a55f1f28edc","resolution":{"observed_at":"2026-05-21T07:09:47.132742Z","resolver_source":"raw_fallback","status":"malformed_identifier"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"0305.359956","doi":"10.1145/3580305.3599561","metadata_source":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"How to dp-fy ml: A practical tutorial to machine learning with differential privacy","venue":null,"work_id":"8fc6a81d-2b67-43d8-a27c-86b4377fc9f7","year":2023},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:6a554dd27bfe6f0c8ed9dd7a6a818829ba081e9bc41d4cd95ea31d321b8a6869","observation_id":"cb67307a-a333-4194-9814-6d1052e6beae","resolution":{"observed_at":"2026-05-21T07:09:46.148156Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-07-09T09:21:37.710816+00:00","source":"crossref_status_cache"},{"observed_at":"2026-07-09T09:21:37.710816+00:00","source":"openalex_status_cache"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2109.12298","last_updated":"2022-08-22T20:24:50Z","snapshot_observed_at":"2026-08-06T15:46:33.397187Z","submitted_at":"2021-09-25T07:10:54Z","title":"Opacus: User-Friendly Differential Privacy Library in PyTorch","version":4},"cited_work":{"arxiv_id":"2109.12298","doi":null,"metadata_source":"pith","pith_arxiv_id":"2109.12298","snapshot_observed_at":"2026-07-08T10:14:52.062226Z","title":"Opacus: User-friendly differential privacy library in pytorch","venue":"cs.LG","work_id":"a7658d3e-1dfe-4593-a474-56151c484a58","year":2021},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"cited_paper":"/paper/2109.12298","citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:492e5329394c114ba83c45acc68f6557e141957cdc379ae92e77d8e65780c341","observation_id":"23578c6f-a480-477f-8674-e1536ec32baa","resolution":{"observed_at":"2026-05-21T07:09:46.326541Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Model-agnostic meta-learning for fast adaptation of deep networks","venue":null,"work_id":"a8f69225-7885-421c-93d4-44b2d19120c9","year":2017},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:6599611297fa17c8a2f40953b991d4ede19824bb120beed1eb640f8e3b4a585b","observation_id":"07a2dcc3-715e-4062-9ccc-dcb80530c7ec","resolution":{"observed_at":"2026-05-21T07:09:47.131011Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-09T07:56:05.250152Z","title":"Gradient-based learning applied to document recognition","venue":null,"work_id":"ae3bda57-6e6d-4f52-8605-aa4b776a2b39","year":1998},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:a3ba169dddc32c3fdc75403afec9621fa8b0d6009bd3a5676e5fd3c7262c9091","observation_id":"5588f9d5-abaf-4c19-b816-a63122271055","resolution":{"observed_at":"2026-05-21T07:09:47.139032Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"The eicu collaborative research database, a freely available multi-center database for critical care research","venue":null,"work_id":"eac833b2-8d5b-4a71-9a52-19afc51f08f3","year":2018},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:1af6c6ac93dc508ac9bf4f74daba4c538b559b9ecd76eb1e15d01515b7b2403c","observation_id":"0ae2dd6c-5f65-40b5-85d4-89b96d399b57","resolution":{"observed_at":"2026-05-21T07:09:47.146711Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Mimic-iv, a freely accessible electronic health record dataset","venue":null,"work_id":"6390521e-6eca-47f4-acf3-ca92f1f2dee0","year":2023},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:710a052eaa00d0d2877c30ab24c51aacad3f213175843485201d1bd592203d4b","observation_id":"7cdf2d71-d96e-43ae-a361-6beeabb7ffa0","resolution":{"observed_at":"2026-05-21T07:09:47.155870Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Introducing the blendedicu dataset, the first harmonized, international intensive care dataset","venue":null,"work_id":"155af7f8-6666-4581-81bb-fefaec82c29e","year":2023},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:b6c5047df289da17494ffd31a51aebc617f03d256ba3dc62329d22d717b95df5","observation_id":"135cc581-21ee-4abd-94cf-cc3018ed521b","resolution":{"observed_at":"2026-05-21T07:09:47.163284Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"An Extensive Data Processing Pipeline for MIMIC-IV","venue":null,"work_id":"7e6abf98-9f24-4247-8294-67930b9c9e3a","year":null},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:05395677494c09c19fa522c277022c8192aff457ed66d416fbb58dd83a56e122","observation_id":"b7bcce44-7202-45a1-bf3f-209d0f29d3ae","resolution":{"observed_at":"2026-05-21T07:09:47.166987Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":null,"venue":null,"work_id":"ed6d3401-9255-426f-9457-4c7c4a442115","year":2022},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:18058594c41211f408772848c108a03af4f49f557255b27298bf0a7f17303e85","observation_id":"653b0688-2dee-4edd-a3cb-07b27045fa12","resolution":{"observed_at":"2026-05-21T07:09:47.124752Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"His research interests include compressed sensing, machine learning, high-dimensional approximations and numerical solution of partial differential equations","venue":null,"work_id":"686c6987-1c1a-4602-bfd8-a4dd463827bf","year":2022},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:c443221e63f1d56c6e3cd844a5a6d1660916e2fcadea55a59c2085dafb014d9c","observation_id":"d439d0a0-11de-47c7-b703-022fa7c7a60b","resolution":{"observed_at":"2026-05-21T07:09:47.126826Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Her research interests include federated learning, differential privacy, synthetic data generation, and distributed optimization","venue":null,"work_id":"8a4a0aca-0049-4c55-8648-152fa20fe3dc","year":2016},"citing_paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees","version":1},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-05-21T07:06:34.274842Z"},"links":{"citing_paper":"/paper/2605.20521"},"observation_digest":"sha256:449049cca9a004f5d3000c98dbebf586df344d6c74512dc96def378ddcc353eb","observation_id":"903bb997-5274-4dc9-8871-cedb9f383eff","resolution":{"observed_at":"2026-05-21T07:09:47.129008Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}}],"paper":{"arxiv_id":"2605.20521","last_updated":"2026-05-19T21:43:22Z","latest_version":1,"primary_category":"cs.LG","snapshot_observed_at":"2026-07-06T23:31:03.878152Z","submitted_at":"2026-05-19T21:43:22Z","title":"An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees"},"reference_resolution":{"displayed":33,"state_counts":{"malformed_identifier":1,"metadata_mismatch":3,"parse_uncertain":0,"unresolved":1,"verified_exact":6,"verified_fuzzy":22},"total_outbound_references":33},"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":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"thesis":"As of 11 August 2026, this Paper Citation Record lists 33 of 33 outbound references and 0 inbound Pith citation observations for arXiv:2605.20521."}