{"as_of":"2026-08-09T22:04:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:09d2acc307923c9436339e465c64e2a569dc656822f6b76c2910b196ab27a037","coverage":[{"denominator":61,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":61,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-01T23:20:45.289168Z","state":"measured"},{"denominator":61,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":61,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-09T06:31:02.800959+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/2607.15467/citation-record","integrity":"/paper/2607.15467/integrity","json":"/paper/2607.15467/citation-record.json","paper":"/paper/2607.15467"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:40.400490Z","title":"Learning complex, extended sequences using the principle of history compression,","venue":null,"work_id":null,"year":1992},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:40.400490Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:29a7148b766229ae305114ce34831284e2b3e8a258d91e54befaef5b43c2b16b","observation_id":"30c13e43-6395-4411-8494-f3ef72853581","resolution":{"observed_at":"2026-08-01T23:20:40.400490Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1503.02531","last_updated":"2015-03-09T15:44:49Z","snapshot_observed_at":"2026-07-06T04:11:24.157003Z","submitted_at":"2015-03-09T15:44:49Z","title":"Distilling the Knowledge in a Neural Network","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1503.02531","snapshot_observed_at":"2026-08-01T23:20:40.434531Z","title":"Distilling the knowledge in a neural network,","venue":null,"work_id":null,"year":2015},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:40.434531Z"},"links":{"cited_paper":"/paper/1503.02531","citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:bbf514d202fb63a6d8fde93b88a5f6e4bea02dccd9e3eec7cdf42da977eaa24f","observation_id":"022411af-9cab-436d-b182-49f5e3f2ca95","resolution":{"observed_at":"2026-08-01T23:20:40.434531Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1704.04861","last_updated":"2017-04-17T03:57:34Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2017-04-17T03:57:34Z","title":"MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1704.04861","snapshot_observed_at":"2026-08-01T23:20:40.478910Z","title":"Mobilenets: Efficient convolutional neural networks for mobile vision applications,","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:40.478910Z"},"links":{"cited_paper":"/paper/1704.04861","citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:eb18435f7fea0c1117cd79f6ca6019fe59f0a5171956dcad9921dbe66003d60b","observation_id":"121e4a97-1c99-42b5-93a3-b6e02df938c2","resolution":{"observed_at":"2026-08-01T23:20:40.478910Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1910.01108","last_updated":"2020-03-01T02:57:50Z","snapshot_observed_at":"2026-08-07T19:07:36.327251Z","submitted_at":"2019-10-02T17:56:28Z","title":"DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1910.01108","snapshot_observed_at":"2026-08-01T23:20:40.543574Z","title":"Distilbert, a distilled version of bert: smaller, faster, cheaper and lighter,","venue":null,"work_id":null,"year":1910},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:40.543574Z"},"links":{"cited_paper":"/paper/1910.01108","citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:943c4fe65177c78b61b2d9f14540f4f67dd9eeda253cdafddea6b98cd490d8bc","observation_id":"39f1de27-ef7c-4173-a5a0-27f0e73c844e","resolution":{"observed_at":"2026-08-01T23:20:40.543574Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:40.592633Z","title":"Training data-efficient image transformers & distillation through attention,","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:40.592633Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:59df9985e89929f6911e5bd2ed8e444307e84bd64d8212cd2b7c14703f039465","observation_id":"9900d72b-d848-442c-a48f-72afce4a3fd9","resolution":{"observed_at":"2026-08-01T23:20:40.592633Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:40.637386Z","title":"Knowledge distillation: A survey,","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:40.637386Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:56aab98caa42f89132c06702c8b7ddf58ad2cce14ef873aca3c4e1612f41e253","observation_id":"40d1e6d1-01a2-4a58-b2ce-0a4a2ba7ded0","resolution":{"observed_at":"2026-08-01T23:20:40.637386Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:40.713343Z","title":"Stealing machine learning models via prediction{APIs},","venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:40.713343Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:3e48025eac703e32d435e00a34830dda591ddce4d4e67d0e4e76c80f12a72806","observation_id":"64470911-9015-408b-8763-8a0b2bbd91bd","resolution":{"observed_at":"2026-08-01T23:20:40.713343Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:40.815008Z","title":"A survey on model extraction attacks and defenses for large language models,","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:40.815008Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:e0a856d3648dfaca7310ead931b60a67deaf8c8ff69388c9db7a4a301d9016e7","observation_id":"2ce9659f-f90a-4ab5-9239-6302c04f5dcc","resolution":{"observed_at":"2026-08-01T23:20:40.815008Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:40.882550Z","title":"DeepSeek trained AI model using distillation, now a dis- ruptive force,","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:40.882550Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:580d01ff0619dba323cab702080ca41b8ab442223ad38620413b08c4b34bc84d","observation_id":"c31efe10-d0c2-426d-984d-272364c728c9","resolution":{"observed_at":"2026-08-01T23:20:40.882550Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:40.987101Z","title":"Detecting and preventing distillation attacks,","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:40.987101Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:82b588f76b72be5bb3799062a3de330b178d059635c90740ab7be67d94426697","observation_id":"3ae36f06-2fe5-47cc-ac3d-e22199e396ed","resolution":{"observed_at":"2026-08-01T23:20:40.987101Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2304.14613","last_updated":"2023-06-17T09:08:18Z","snapshot_observed_at":"2026-07-06T15:21:04.733820Z","submitted_at":"2023-04-28T03:34:43Z","title":"Deep Intellectual Property Protection: A Survey","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2304.14613","snapshot_observed_at":"2026-08-01T23:20:41.181212Z","title":"Deep intellectual property protection: A survey,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:41.181212Z"},"links":{"cited_paper":"/paper/2304.14613","citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:d47d2ebfdc6feec67b3dee6b56bccb87138b05a37b56694e95353836af97cf16","observation_id":"e209b7e1-bcc3-441e-a1e5-24811f375a8f","resolution":{"observed_at":"2026-08-01T23:20:41.181212Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:41.316028Z","title":"Intellectual property protection for deep learning model and dataset intelligence,","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:41.316028Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:436ec96fdcbc83a2a4a15870db5653ae271c06ae0021c9831efdb7554451deb4","observation_id":"33e9605a-17da-430b-aee6-46193756718a","resolution":{"observed_at":"2026-08-01T23:20:41.316028Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:41.399828Z","title":"Membership inference attacks against machine learning models,","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:41.399828Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:b1e9d45af30b7cf4c001c02a7161a62e79b44c3bcd88d7779dec5e772b9f7b05","observation_id":"0a02e48e-6904-48e8-ab5e-d0883030aa49","resolution":{"observed_at":"2026-08-01T23:20:41.399828Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1806.01246","last_updated":"2018-12-14T19:39:43Z","snapshot_observed_at":"2026-08-08T03:39:28.408433Z","submitted_at":"2018-06-04T17:38:42Z","title":"ML-Leaks: Model and Data Independent Membership Inference Attacks and Defenses on Machine Learning Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1806.01246","snapshot_observed_at":"2026-08-01T23:20:41.497018Z","title":"Ml-leaks: Model and data independent membership inference at- tacks and defenses on machine learning models,","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:41.497018Z"},"links":{"cited_paper":"/paper/1806.01246","citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:ff963d499b5680db64aa0332c696e13c7a49bae5b3c06b374d6eb1af2c20cd96","observation_id":"2f990f9f-6242-4cc5-b7bd-1b0790782f9c","resolution":{"observed_at":"2026-08-01T23:20:41.497018Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:41.572075Z","title":"Practical black-box attacks against machine learning,","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:41.572075Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:823332bf3df6766666b2d50569a62125ee9a31068dbc85f74022acba7c2eab29","observation_id":"2b54a3f9-226b-4772-bb19-a52787462cfc","resolution":{"observed_at":"2026-08-01T23:20:41.572075Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:41.702915Z","title":"Evasion attacks against machine learning at test time,","venue":null,"work_id":null,"year":2013},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:41.702915Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:69e45ca12e6bf38638c73843accba6f8a06ef52ac56a3c4e8a7d85cd71470e5b","observation_id":"ebc8e058-bdee-4ea2-ae6d-8c86ade158a8","resolution":{"observed_at":"2026-08-01T23:20:41.702915Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:41.799827Z","title":"Adversarial learning,","venue":null,"work_id":null,"year":2005},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:41.799827Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:c8941ad90fd3adff110e7d4bde0c6a758e9c1b67da247cfb0b7856d8f03a63ed","observation_id":"424b66ac-0242-4f9f-b298-ed46b5f15770","resolution":{"observed_at":"2026-08-01T23:20:41.799827Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:41.905545Z","title":"Privacy in pharmacogenetics: An{End-to-End}case study of person- alized warfarin dosing,","venue":null,"work_id":null,"year":2014},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:41.905545Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:723e91c37a6da47cf3bce90179efde0f6891859b463109f3defd7f0b01432cdb","observation_id":"783a80a5-acb3-451f-804a-175ce3f2cae1","resolution":{"observed_at":"2026-08-01T23:20:41.905545Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:41.957138Z","title":"Model inversion attacks that exploit confidence information and basic countermeasures,","venue":null,"work_id":null,"year":2015},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:41.957138Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:f5c11c39f400ca4bfbcf227a8af3f281e269844e13b7b61d9ef358c68e242f3a","observation_id":"75d44baf-be61-49b3-9951-5a62673cf0df","resolution":{"observed_at":"2026-08-01T23:20:41.957138Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:42.002760Z","title":"Hacking smart machines with smarter ones: How to extract meaningful data from machine learning classifiers,","venue":null,"work_id":null,"year":2015},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:42.002760Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:1cc2b371620c8ba5ad58ef2e305ae4646fb981df043753bf01bf87c48a173b05","observation_id":"3e11b25b-a309-45e2-80d4-56db5bce3972","resolution":{"observed_at":"2026-08-01T23:20:42.002760Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:42.045930Z","title":"Artificial intelligence and machine learn- ing in clinical medicine, 2023,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:42.045930Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:edc7ad256396593103ab332356a29c9d5052a9a0305e0adc83f72404bc23d5fd","observation_id":"9b727c59-e84a-475b-8ebf-bc587c02abf8","resolution":{"observed_at":"2026-08-01T23:20:42.045930Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:42.082555Z","title":"Financial fraud detection through the application of machine learning techniques: a literature review,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:42.082555Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:8772b13202c2e253fbc4b7eb64733248cb73a7a7ae370729de5c99b40c049754","observation_id":"aec2fe46-760f-4fc8-bd01-c845d786dea1","resolution":{"observed_at":"2026-08-01T23:20:42.082555Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:42.142365Z","title":"Learn- ing lightweight object detectors via multi-teacher progressive distilla- tion,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:42.142365Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:c8726b1b5f8719fbad7c3cdcfd6f7474e3439b1b1023748f895b803ee5b05561","observation_id":"aa414ec0-1eeb-4a9f-b592-76105a8829fb","resolution":{"observed_at":"2026-08-01T23:20:42.142365Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:42.208161Z","title":"Antidistillation sampling,","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:42.208161Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:38b6d8d9792bc41660a5cc27ad5074bc24c4aa21d86e68b9f16783e255e15149","observation_id":"ec44827c-915e-427a-a549-7ebbb6b79287","resolution":{"observed_at":"2026-08-01T23:20:42.208161Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:42.253168Z","title":"Defending against neural network model stealing attacks using deceptive perturbations,","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:42.253168Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:8037c9777bf49145f5b7fba814077cd47859a35107574501fdaae1fe1b0980cc","observation_id":"b521e6b0-85d4-4668-b93d-c887b48f109a","resolution":{"observed_at":"2026-08-01T23:20:42.253168Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:42.311615Z","title":"D-dae: Defense- penetrating model extraction attacks,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:42.311615Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:ac2f9f672b4a24f96d6a2b0e205e0e96be9529aaf15f8aac11406a21fcb150b8","observation_id":"468f5835-5c16-4177-849a-cff0841fe820","resolution":{"observed_at":"2026-08-01T23:20:42.311615Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:42.372283Z","title":"On calibration of modern neural networks,","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:42.372283Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:8ab376dab00aac1129a09108e22a9de4256565a31e1c2ffd357de8dfffde11c7","observation_id":"7b5a7eb8-6bd0-4988-aef5-9055bc7ce3a5","resolution":{"observed_at":"2026-08-01T23:20:42.372283Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:42.426803Z","title":"Model compres- sion,","venue":null,"work_id":null,"year":2006},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:42.426803Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:52a2630e122fb0d1f8585b3e2215646e5b69e91aeb28bc6085e43b740a09b5e1","observation_id":"8e7e4002-a315-4c2f-b94a-2f6d75b09816","resolution":{"observed_at":"2026-08-01T23:20:42.426803Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:42.499397Z","title":"Do deep nets really need to be deep?","venue":null,"work_id":null,"year":2014},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:42.499397Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:f2d51c6c845352979855ac7d29695bf7d206c661d7f05cf25ee0467c075773c2","observation_id":"4b622548-0c5a-4a97-8884-2f52c96d65c7","resolution":{"observed_at":"2026-08-01T23:20:42.499397Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:42.574438Z","title":"Born again neural networks,","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:42.574438Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:27ecd819248439b856ed284c77d0935d6d76bf3c63d1945a91708391fb2aa8d7","observation_id":"3017901f-abc9-467f-9d76-a72275f0e310","resolution":{"observed_at":"2026-08-01T23:20:42.574438Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2002.03532","last_updated":"2021-02-28T23:31:44Z","snapshot_observed_at":"2026-07-06T08:55:58.482745Z","submitted_at":"2020-02-10T04:21:41Z","title":"Understanding and Improving Knowledge Distillation","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2002.03532","snapshot_observed_at":"2026-08-01T23:20:42.682110Z","title":"Understanding and improving knowledge distillation,","venue":null,"work_id":null,"year":2002},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:42.682110Z"},"links":{"cited_paper":"/paper/2002.03532","citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:29cddb4bbe03006020d72402b45892623b2b473c420291c59f5d0f4a708a3f1b","observation_id":"ada97991-d776-4feb-bb34-3f307bdde3db","resolution":{"observed_at":"2026-08-01T23:20:42.682110Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:42.769138Z","title":"Knockoff nets: Stealing functionality of black-box models,","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:42.769138Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:2552f3cf703fb40e34cf61d3366dd4566b776f02ee8b181856756ed4bf53088e","observation_id":"0c0112db-ccd3-4751-a2ca-86ce676669fd","resolution":{"observed_at":"2026-08-01T23:20:42.769138Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:42.839606Z","title":"Data-free model extraction,","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:42.839606Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:cca0abe89cbf205385962643bf9fd3ba93f139ce4889de65e336fb6fb5385ad4","observation_id":"0f71540d-ee4b-40a0-b278-ede6942eca09","resolution":{"observed_at":"2026-08-01T23:20:42.839606Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:43.003743Z","title":"Black-box behavioral distillation breaks safety alignment in medical llms,","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:43.003743Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:7e42bb143f2f6c05fa6a2bc3e09d3a8ea26f931583b6cd6c08b12c21de49c866","observation_id":"9a063281-9845-4c4e-b993-70d67cdf617b","resolution":{"observed_at":"2026-08-01T23:20:43.003743Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:43.123367Z","title":"I know what you trained last summer: A survey on stealing machine learning models and defences,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:43.123367Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:bf9b9b2be31ddf94878fe49a0cf6ed5ab4149264f0138bc902f006d941f75674","observation_id":"653030e9-c9fa-4c31-9e88-1ec1ae7cd72a","resolution":{"observed_at":"2026-08-01T23:20:43.123367Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:43.198987Z","title":"A comprehensive de- fense framework against model extraction attacks,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:43.198987Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:ebe01ceb992975202839da7d872300699d98b6fc09ea41e65777858e00a9fe0c","observation_id":"6c82d18e-88c6-4fa3-9c72-01f2b97de085","resolution":{"observed_at":"2026-08-01T23:20:43.198987Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2402.13116","last_updated":"2024-10-21T16:22:33Z","snapshot_observed_at":"2026-08-02T17:17:56.296462Z","submitted_at":"2024-02-20T16:17:37Z","title":"A Survey on Knowledge Distillation of Large Language Models","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2402.13116","snapshot_observed_at":"2026-08-01T23:20:43.259453Z","title":"A survey on knowledge distillation of large language models,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:43.259453Z"},"links":{"cited_paper":"/paper/2402.13116","citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:544bd1c2c4f5515bf2573e0cb00439ec1e8cf526ca516426c6f06145cedc749b","observation_id":"c2bf0f51-4c38-48b4-99b8-36ea4c10a2e2","resolution":{"observed_at":"2026-08-01T23:20:43.259453Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2105.07381","last_updated":"2021-05-16T08:41:30Z","snapshot_observed_at":"2026-07-06T11:09:48.217065Z","submitted_at":"2021-05-16T08:41:30Z","title":"Undistillable: Making A Nasty Teacher That CANNOT teach students","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2105.07381","snapshot_observed_at":"2026-08-01T23:20:43.312555Z","title":"Undistillable: Making a nasty teacher that cannot teach students,","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:43.312555Z"},"links":{"cited_paper":"/paper/2105.07381","citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:73d6d0f489b87d427f386b002f8fd1933dffc777c0d61d5330e0d848c323b14a","observation_id":"f0078587-749b-4b33-81dc-53a258f5b9fc","resolution":{"observed_at":"2026-08-01T23:20:43.312555Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:43.368941Z","title":"Adversarial sparse teacher: Defense against distillation-based model stealing attacks using adversarial examples,","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:43.368941Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:3e5d61a46c2b935d49ca7cfd078cb867fb640a6ae95b7de02fe8e3c19d5f4dac","observation_id":"0d407a49-b5c6-4a81-b006-3c0def660f26","resolution":{"observed_at":"2026-08-01T23:20:43.368941Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:43.445415Z","title":"Defending against model extraction attacks with ood feature learning and decision boundary confusion,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:43.445415Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:b1d74bb01f40b73b19429473f9581b592f0c8e4d305039554d9de3e78accb82c","observation_id":"0fb3e5e6-d761-4b21-8d45-5d9f5db41252","resolution":{"observed_at":"2026-08-01T23:20:43.445415Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2506.02362","last_updated":"2025-06-03T01:37:09Z","snapshot_observed_at":"2026-08-09T15:19:23.285991Z","submitted_at":"2025-06-03T01:37:09Z","title":"MISLEADER: Defending against Model Extraction with Ensembles of Distilled Models","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2506.02362","snapshot_observed_at":"2026-08-01T23:20:43.484131Z","title":"Misleader: Defending against model extraction with ensembles of distilled models,","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:43.484131Z"},"links":{"cited_paper":"/paper/2506.02362","citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:3645ee4f04f5078123bcaef31502a494e288b7446583cfb00104318f7f37e357","observation_id":"9dde8944-2c60-4c02-844d-c0fd23468225","resolution":{"observed_at":"2026-08-01T23:20:43.484131Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:43.570463Z","title":"Prada: protecting against dnn model stealing attacks,","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:43.570463Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:046713cdb804b8ccc8d4c73da6402d1d571cce1c0333605ec0737fea25765438","observation_id":"aff6aa58-31f0-4d79-9018-3698488dfcc6","resolution":{"observed_at":"2026-08-01T23:20:43.570463Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2403.18580","last_updated":"2024-03-27T13:59:21Z","snapshot_observed_at":"2026-07-06T17:51:56.621426Z","submitted_at":"2024-03-27T13:59:21Z","title":"MisGUIDE : Defense Against Data-Free Deep Learning Model Extraction","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2403.18580","snapshot_observed_at":"2026-08-01T23:20:43.657186Z","title":"Misguide: Defense against data-free deep learning model extraction,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:43.657186Z"},"links":{"cited_paper":"/paper/2403.18580","citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:50538b0db72856a96cf311711b52d01659828c9311518c837a9a66b24aa23599","observation_id":"0617338a-0c30-436a-9cbe-8d65401eb0ec","resolution":{"observed_at":"2026-08-01T23:20:43.657186Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:43.706438Z","title":"Radep: A resilient adaptive defense framework against model extraction attacks,","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":44,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:43.706438Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:39a1d9e7855dcf23928f96e673c839e4f18b00b08b02115596e57476756c12d1","observation_id":"7d4cdddf-cab5-4118-b235-a7829775fe0d","resolution":{"observed_at":"2026-08-01T23:20:43.706438Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:43.765920Z","title":"The Sybil attack,","venue":null,"work_id":null,"year":2002},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":45,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:43.765920Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:8cf8b3d04979a38459021aac23e6f743a033f1c7b97e4f54e59d5755b7937c48","observation_id":"32f0f09a-03e6-416e-82f4-dda1af74a79e","resolution":{"observed_at":"2026-08-01T23:20:43.765920Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:43.831799Z","title":"Cloudleak: Large-scale deep learning models stealing through adversarial exam- ples","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":46,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:43.831799Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:22bed3bccf5b9ec4e65a95a1d694b91f5d50a76349820cee58d213e681cf4a28","observation_id":"1ef47050-e549-4255-ac3f-5e9be2c0abf1","resolution":{"observed_at":"2026-08-01T23:20:43.831799Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:43.901881Z","title":"Embedding water- marks into deep neural networks,","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":47,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:43.901881Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:5138997fdc1b89aa21bbe10baa63f653a6699bcc28c8004950540f1925732fe7","observation_id":"8427acde-3eb8-4f94-ba0c-d0dcf8551354","resolution":{"observed_at":"2026-08-01T23:20:43.901881Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:43.957485Z","title":"Entangled watermarks as a defense against model extraction,","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":48,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:43.957485Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:5b0245977d3c019f037de4d2742b50649c48e9bdaa9e516edd98a4ab97590802","observation_id":"aa6abbdf-3cc1-426f-86de-54bbc26a39b7","resolution":{"observed_at":"2026-08-01T23:20:43.957485Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:44.032840Z","title":"Dawn: Dynamic adversarial watermarking of neural networks,","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":49,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:44.032840Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:6040aa166fb8e2556d22652264f3271d11d2262d20be7316696256fb97f4cc17","observation_id":"69355339-1d87-40f0-b8aa-894f5c2816d5","resolution":{"observed_at":"2026-08-01T23:20:44.032840Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:44.110544Z","title":"Defense against model extraction attack by bayesian active watermarking,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":50,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:44.110544Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:1b69ef6133d53602d19da3b34041980190eb62fa67a939bbccdafac462ea8d03","observation_id":"8778175c-64ac-489f-a60a-557f23fd1407","resolution":{"observed_at":"2026-08-01T23:20:44.110544Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1906.10908","last_updated":"2020-03-03T10:51:12Z","snapshot_observed_at":"2026-08-07T17:53:53.891684Z","submitted_at":"2019-06-26T08:32:37Z","title":"Prediction Poisoning: Towards Defenses Against DNN Model Stealing Attacks","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1906.10908","snapshot_observed_at":"2026-08-01T23:20:44.211113Z","title":"Prediction poisoning: To- wards defenses against dnn model stealing attacks,","venue":null,"work_id":null,"year":1906},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":51,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:44.211113Z"},"links":{"cited_paper":"/paper/1906.10908","citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:c3019a37598eaf518dc03cbded7c287dbbc6ba787c1353ff3544d292a7d64839","observation_id":"ae3fd64b-c402-48bc-bb9b-3970dddf2595","resolution":{"observed_at":"2026-08-01T23:20:44.211113Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:44.317864Z","title":"{ModelGuard}:{Information-Theoretic}defense against model extraction attacks,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":52,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:44.317864Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:4b8e214fa31bb732f63878b3dfa64422e9b53953ad36bf37aff8e8e582986a4c","observation_id":"f28e6b34-7b38-437e-b035-7b83d1fc08c1","resolution":{"observed_at":"2026-08-01T23:20:44.317864Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:44.412260Z","title":"Efficient model stealing defense with noise transition matrix,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":53,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:44.412260Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:916425e24b0cf3ec315fabc8c23f4b4d1b11a097e2011258122268751ba4de7d","observation_id":"40223032-1bdd-43ae-9b94-ba0732077b13","resolution":{"observed_at":"2026-08-01T23:20:44.412260Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2602.03396","last_updated":"2026-05-06T12:26:57Z","snapshot_observed_at":"2026-07-06T22:44:18.875334Z","submitted_at":"2026-02-03T11:16:59Z","title":"Towards Distillation-Resistant Large Language Models: An Information-Theoretic Perspective","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2602.03396","snapshot_observed_at":"2026-08-01T23:20:44.486147Z","title":"Towards distillation-resistant large lan- guage models: An information-theoretic perspective,","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":54,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:44.486147Z"},"links":{"cited_paper":"/paper/2602.03396","citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:9f9d8bdb02a1ad51eefd49ffc3d3f1d333b0600551a7ff3c0dc6be7408b90924","observation_id":"846a3e29-f40e-4e7f-86bf-38238ced16d9","resolution":{"observed_at":"2026-08-01T23:20:44.486147Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1701.06548","last_updated":"2017-01-23T18:35:28Z","snapshot_observed_at":"2026-08-05T00:39:48.314299Z","submitted_at":"2017-01-23T18:35:28Z","title":"Regularizing Neural Networks by Penalizing Confident Output Distributions","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1701.06548","snapshot_observed_at":"2026-08-01T23:20:44.623823Z","title":"Regu- larizing neural networks by penalizing confident output distributions,","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":55,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:44.623823Z"},"links":{"cited_paper":"/paper/1701.06548","citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:f05d508aaa1a8d0306e7c0ee69a8ad24cd8c7a45c0fc95db49231d68afeaeb4b","observation_id":"78af3e37-fbf5-486c-96b9-f4205ec2a98b","resolution":{"observed_at":"2026-08-01T23:20:44.623823Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:44.774490Z","title":"Adaptive temperature scal- ing for robust calibration of deep neural networks,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":56,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:44.774490Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:3c28760879896af1b7fe03d1165ac5b67b2481b4fe59a09f8002b461d6c07b2f","observation_id":"96e54412-a669-4534-9c4c-f67624682b04","resolution":{"observed_at":"2026-08-01T23:20:44.774490Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:44.869095Z","title":"Sample- dependent adaptive temperature scaling for improved calibration,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":57,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:44.869095Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:bf30c38844708aa170d689fbd55519592e804c97cdc9b77b448c9d89e4bcb93a","observation_id":"bb7c92c0-315b-4134-9362-7e537d947e4b","resolution":{"observed_at":"2026-08-01T23:20:44.869095Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1810.11586","last_updated":"2019-05-08T18:51:05Z","snapshot_observed_at":"2026-07-06T07:10:55.945155Z","submitted_at":"2018-10-27T03:03:57Z","title":"Attended Temperature Scaling: A Practical Approach for Calibrating Deep Neural Networks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1810.11586","snapshot_observed_at":"2026-08-01T23:20:44.982351Z","title":"Attended temperature scaling: a practical approach for calibrating deep neural networks,","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":58,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:44.982351Z"},"links":{"cited_paper":"/paper/1810.11586","citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:bcda73377c82aba887c1bdeb120637c2ed026471ecf2e089a8b3a542f797459c","observation_id":"a1789fe2-3669-4317-945f-61b7b53b8766","resolution":{"observed_at":"2026-08-01T23:20:44.982351Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:45.147709Z","title":"Asymmetric temperature scaling makes larger networks teach well again,","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":59,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:45.147709Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:9e9c28591aff74fee1786a6a44ddd408420d2e1103a675561356be5984c40593","observation_id":"20ff2fbd-8bb9-4a01-bc01-91d6ddea68cf","resolution":{"observed_at":"2026-08-01T23:20:45.147709Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:45.289168Z","title":"Research computing services,","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":60,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:45.289168Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:ad3400f3addc8997ad4760a9147bb5bf5736663f9230f3e3de0502130093a0db","observation_id":"e75914da-c95e-487b-b60c-4bde9ae16387","resolution":{"observed_at":"2026-08-01T23:20:45.289168Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T23:20:41.079936Z","title":"Available: https://www.anthropic.com/news/ detecting-and-preventing-distillation-attacks","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification","version":1},"reference_index":2026,"source":"pdf_text","source_observed_at":"2026-08-01T23:20:41.079936Z"},"links":{"citing_paper":"/paper/2607.15467"},"observation_digest":"sha256:e988b8143121ca912e1fade3de221905a8e5a98b1452481c438ba50afd355061","observation_id":"36e8e748-33aa-42f4-b410-488acedd8191","resolution":{"observed_at":"2026-08-01T23:20:41.079936Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"paper":{"arxiv_id":"2607.15467","last_updated":"2026-07-16T21:33:37Z","latest_version":1,"primary_category":"cs.LG","snapshot_observed_at":"2026-08-06T15:56:21.552313Z","submitted_at":"2026-07-16T21:33:37Z","title":"ADS-C: Antidistillation Sampling for Classification"},"reference_resolution":{"displayed":61,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":61,"verified_exact":0,"verified_fuzzy":0},"total_outbound_references":61},"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":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+00:00","source":"retraction_watch"}],"thesis":"As of 9 August 2026, this Paper Citation Record lists 61 of 61 outbound references and 0 inbound Pith citation observations for arXiv:2607.15467."}