{"as_of":"2026-08-09T03:55:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:bfe28129de43469c5d07eb26b92722bda350acecc5a486cec6a2f726dfaa3a12","coverage":[{"denominator":46,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":46,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-08T13:30:03.623879Z","state":"measured"},{"denominator":46,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":46,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-08T06:32:00.761636+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/2502.07231/citation-record","integrity":"/paper/2502.07231/integrity","json":"/paper/2502.07231/citation-record.json","paper":"/paper/2502.07231"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:04.201756Z","title":"Past, present, and future of face recognition: A review","venue":null,"work_id":"89435fb8-81a0-41a4-9a02-15d0587e58de","year":2020},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.447888Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:b7f8718d3cf14e590108752a7b6ff8555943f69964a19531155fa773fc358130","observation_id":"e145fcf4-c693-4183-b067-5a458b967448","resolution":{"observed_at":"2026-08-08T13:30:04.206073Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2203.17274","last_updated":"2022-06-03T17:52:04Z","snapshot_observed_at":"2026-07-06T12:55:27.843060Z","submitted_at":"2022-03-31T17:59:30Z","title":"Exploring Visual Prompts for Adapting Large-Scale Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2203.17274","snapshot_observed_at":"2026-08-08T13:30:03.452514Z","title":"Exploring visual prompts for adapting large-scale models","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.452514Z"},"links":{"cited_paper":"/paper/2203.17274","citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:c4332d8a226344def90685837b82eb1b41568e90bd376c2d4e9f7b54944bbb22","observation_id":"daa4f87e-5307-4586-b804-5939b38108c3","resolution":{"observed_at":"2026-08-08T13:30:03.452514Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:04.189382Z","title":"A new backdoor attack in cnns by training set corruption without label poisoning","venue":null,"work_id":"16ac2b79-cc80-4560-9618-9e53e41ca45d","year":2019},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.457334Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:72b8c7a6157dcb07ecd2f952eea21cf420964c62c9e5582dfaa4f609bf38ed49","observation_id":"fa8a2b4a-92da-44ff-9ca5-96cff2e23a47","resolution":{"observed_at":"2026-08-08T13:30:04.193800Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1809.11096","last_updated":"2019-02-25T21:32:06Z","snapshot_observed_at":"2026-07-06T07:04:57.275371Z","submitted_at":"2018-09-28T15:38:49Z","title":"Large Scale GAN Training for High Fidelity Natural Image Synthesis","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1809.11096","snapshot_observed_at":"2026-08-08T13:30:03.461944Z","title":"Large scale gan training for high fidelity natural image synthesis","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.461944Z"},"links":{"cited_paper":"/paper/1809.11096","citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:8196001dc61f3acf984a485b4e71773f0143ee9ed3fa233866a24bbcd972fbf6","observation_id":"a6ddd7df-6f39-416f-a2e6-c324c764b774","resolution":{"observed_at":"2026-08-08T13:30:03.461944Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:04.177405Z","title":"One-shot neural backdoor erasing via adversarial weight masking","venue":null,"work_id":"b0726c20-35b7-4234-b714-3778c68cb3fe","year":2022},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.466324Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:e8bd4c48fb8d32c92a594c49d50a58bab1127076db73987253c727f534257c62","observation_id":"bb63554e-c083-4761-ae60-83d90077f791","resolution":{"observed_at":"2026-08-08T13:30:04.181308Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:04.165741Z","title":"Detecting backdoor attacks on deep neural networks by activation clustering","venue":null,"work_id":"a1a91d84-54d1-407a-b9a1-66aad3b1c797","year":2019},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.470343Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:68298244c1947f7f757cd7ef3080f1f7ad6a61b9a703237018a35ad344618717","observation_id":"40c921b6-ed03-48e1-8e4d-dc5cf7abc5dc","resolution":{"observed_at":"2026-08-08T13:30:04.169812Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:04.153794Z","title":"Targeted backdoor attacks on deep learning systems using data poisoning","venue":null,"work_id":"40e38cba-e045-4512-8f0d-9de2b4c5584c","year":2017},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.474688Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:cfc036fb823728620e97f0336fac684dc5ec54c5384615bba02c045d932239e7","observation_id":"36478c83-19ab-43b8-aa03-0aaf1d17d59c","resolution":{"observed_at":"2026-08-08T13:30:04.157800Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1810.03505","last_updated":"2018-10-02T21:20:09Z","snapshot_observed_at":"2026-08-07T01:38:08.346703Z","submitted_at":"2018-10-02T21:20:09Z","title":"CINIC-10 is not ImageNet or CIFAR-10","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1810.03505","snapshot_observed_at":"2026-08-08T13:30:03.478578Z","title":"Cinic-10 is not imagenet or cifar-10","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.478578Z"},"links":{"cited_paper":"/paper/1810.03505","citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:0260040b0ae105450525d0b6f372230db5ba824bce062f3d0866aebdc49c7634","observation_id":"b3031070-5302-496e-86d3-1218e521e300","resolution":{"observed_at":"2026-08-08T13:30:03.478578Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2411.11200","last_updated":"2025-01-07T00:23:43Z","snapshot_observed_at":"2026-07-06T19:51:41.646058Z","submitted_at":"2024-11-17T23:30:01Z","title":"Countering Backdoor Attacks in Image Recognition: A Survey and Evaluation of Mitigation Strategies","version":2},"cited_work":{"arxiv_id":"2411.11200","doi":null,"metadata_source":"pith","pith_arxiv_id":"2411.11200","snapshot_observed_at":"2026-08-08T13:30:03.722352Z","title":"Countering Backdoor Attacks in Image Recognition: A Survey and Evaluation of Mitigation Strategies","venue":"cs.CR","work_id":"cd7a40a0-c2a0-4233-8012-fd337508daf5","year":2024},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.482646Z"},"links":{"cited_paper":"/paper/2411.11200","citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:dea773f941a9dd0791ff6f1c45f0cffb0aba93b9f66a2287b3acdbe82d32d4cd","observation_id":"120f2420-d0f7-4da3-ad19-ca97ec43e97c","resolution":{"observed_at":"2026-08-08T13:30:03.726659Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:04.141719Z","title":"Badnets: Evaluating backdooring attacks on deep neural networks","venue":null,"work_id":"f0d00b39-be8f-40a9-befa-a7f9558951d3","year":2019},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.486751Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:5ffbb76e4407a1e0491aae6efb2e1cc45a4923104fb643de9b8a0ad69ad0e322","observation_id":"c14bd710-40db-4d2c-94e9-9d19b86db87b","resolution":{"observed_at":"2026-08-08T13:30:04.145946Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:03.490419Z","title":"Deep residual learning for image recognition","venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.490419Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:90cace884b700c46ab6afa4b24f7703d77b37ddd8723b0912c3c8f93438889b9","observation_id":"11d5899f-cbf6-478d-a1ea-b4e112f9d745","resolution":{"observed_at":"2026-08-08T13:30:03.490419Z","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":"2026-08-08T13:30:03.494448Z","title":"Identity mappings in deep residual networks","venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.494448Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:b90dddb49d61420a5ce5af56fa0f2dbc296443b525062380eca73ef328b668c0","observation_id":"20e936c8-5685-46f3-b025-3e201112fff1","resolution":{"observed_at":"2026-08-08T13:30:03.494448Z","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":"2026-08-08T13:30:03.498023Z","title":"Denoising diffusion probabilistic models","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.498023Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:ae7461dc9884f05f16013ae441d7b010231135a5a3c93a309e62c8c7014a9ac4","observation_id":"74a91be4-4139-49c4-9cbb-53bb035ba06f","resolution":{"observed_at":"2026-08-08T13:30:03.498023Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:04.107274Z","title":"Revisiting data- free knowledge distillation with poisoned teachers","venue":null,"work_id":"d731ffc2-1a0d-48f4-8b7b-d4f37bb55742","year":2023},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.501708Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:e1c3ae4036cf5f32a64f765f900f99e3419cec6376bad810faff8f01baa9d2d6","observation_id":"4c342554-6767-47a3-a560-5257fbce2eec","resolution":{"observed_at":"2026-08-08T13:30:04.111889Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:03.505650Z","title":"Learning multiple layers of features from tiny images","venue":null,"work_id":null,"year":2009},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.505650Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:1a543c2d0d2f9c96ede0454b217c07d7299702e4e10c0e581f05860119d0813e","observation_id":"69859d76-a513-4bfb-a770-6fc81021499c","resolution":{"observed_at":"2026-08-08T13:30:03.505650Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:04.088204Z","title":"Tiny imagenet visual recognition challenge","venue":null,"work_id":"2b8e6cc1-2255-4db7-b979-1a4425b39801","year":2015},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.509328Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:827ec702d975544681c5de5229822d7f2e7d73461a1817645689457d1b02f93b","observation_id":"3d4f3218-11b8-4f64-a175-45bc7097ae73","resolution":{"observed_at":"2026-08-08T13:30:04.092051Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:04.076411Z","title":"Neural attention distillation: Erasing backdoor triggers from deep neural networks","venue":null,"work_id":"b55e5246-f059-4b21-a5ca-769b7ce5f454","year":2021},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.512949Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:7a0a1694f19d64b29cb25d05d6309981b2ef66c0ae12adecca910b617f26aa97","observation_id":"64adac47-dc99-435c-b7f2-5dc3d3a36faa","resolution":{"observed_at":"2026-08-08T13:30:04.080441Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:04.063123Z","title":"Reconstructive neuron pruning for backdoor defense","venue":null,"work_id":"88e65ccb-9e5f-451f-979c-d2d7b3ef63aa","year":2023},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.516669Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:398028e675377af0310dca3a8e158c2cbae74644c660111f22f8319e5620aa75","observation_id":"d2ed2dfe-ce07-4c49-aafe-5bfe9c09fbf0","resolution":{"observed_at":"2026-08-08T13:30:04.067647Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:04.050436Z","title":"Invisible backdoor attack with sample-specific triggers","venue":null,"work_id":"be6eb286-bb18-4472-8d7c-e9700d25fe2c","year":2021},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.520193Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:a7648e24874648e21d88f5bd5f14772acc421b93b4361b4f257cc03f3916d14c","observation_id":"160ab874-1063-4108-aa2f-572ee11ee73a","resolution":{"observed_at":"2026-08-08T13:30:04.054665Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2408.15861","last_updated":"2024-08-28T15:21:10Z","snapshot_observed_at":"2026-07-06T19:07:11.892165Z","submitted_at":"2024-08-28T15:21:10Z","title":"Fusing Pruned and Backdoored Models: Optimal Transport-based Data-free Backdoor Mitigation","version":1},"cited_work":{"arxiv_id":"2408.15861","doi":null,"metadata_source":"pith","pith_arxiv_id":"2408.15861","snapshot_observed_at":"2026-08-08T13:30:03.705861Z","title":"Fusing Pruned and Backdoored Models: Optimal Transport-based Data-free Backdoor Mitigation","venue":"cs.CR","work_id":"d6ae6fbc-0935-45fa-b7b7-de14fde3fac3","year":2024},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.524383Z"},"links":{"cited_paper":"/paper/2408.15861","citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:cb75f15997df4f2a5b4aebec6a0af40eb4d5318c102be21aeaabe5586d279721","observation_id":"0d734157-2f59-4ed9-a2f0-71194e7d5945","resolution":{"observed_at":"2026-08-08T13:30:03.710465Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:04.038339Z","title":"Fine-pruning: Defending against backdooring attacks on deep neural networks","venue":null,"work_id":"67524ef2-d562-418c-8392-ec977efb76aa","year":2018},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.528924Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:aa5320453c5ec40ca1eca08afda99d5415620227aabf0d48d83d43ed6ffe22c9","observation_id":"d134961a-5f8f-4c10-871d-fb365374c80f","resolution":{"observed_at":"2026-08-08T13:30:04.042413Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:04.024053Z","title":"Computing systems for autonomous driving: State of the art and challenges","venue":null,"work_id":"ffcf3b22-43e2-4744-b108-89211fa29a54","year":2020},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.532649Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:80bdcba283180ea954884584bb70501c91d5f765e5104d0e2bab912d68eee323","observation_id":"317fcc12-153e-4fda-a3d1-523f29af8aa5","resolution":{"observed_at":"2026-08-08T13:30:04.028526Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2310.01875","last_updated":"2023-10-21T12:37:05Z","snapshot_observed_at":"2026-07-06T16:26:58.234941Z","submitted_at":"2023-10-03T08:25:32Z","title":"Towards Stable Backdoor Purification through Feature Shift Tuning","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2310.01875","snapshot_observed_at":"2026-08-08T13:30:03.536459Z","title":"Towards stable backdoor purification through feature shift tuning","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.536459Z"},"links":{"cited_paper":"/paper/2310.01875","citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:e6098abf0da6feed4e00e1921ec6e9145371e4d89f38186f60b83d30e4000208","observation_id":"204ace4a-d1c8-45be-a722-bb9511b29e50","resolution":{"observed_at":"2026-08-08T13:30:03.536459Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:04.009347Z","title":"Input-aware dynamic backdoor attack","venue":null,"work_id":"45681993-59d2-47ad-94e7-98d444d94bf1","year":2020},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.540464Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:2cf4ed2d38ee69e9698ec5e853f8de866baf3e170c521c8cf90a3dda966d6b31","observation_id":"5659dd52-bc61-423a-9703-a381d93de444","resolution":{"observed_at":"2026-08-08T13:30:04.014227Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:03.995374Z","title":"Wanet - imperceptible warping-based backdoor attack","venue":null,"work_id":"d7ebbfd4-dd92-426b-a31f-67aa3f3bcc8c","year":2021},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.544297Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:a04cfd0d946e3bbea3bcaec3ba9b15e060a42800ce33f05786c4f9854949bd39","observation_id":"c06fd3d1-789b-49d6-8816-ac679eaecf8f","resolution":{"observed_at":"2026-08-08T13:30:03.999614Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:03.547936Z","title":"Imagenet large scale visual recognition challenge","venue":null,"work_id":null,"year":2015},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.547936Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:3251507af2fa8d0559cde8bd1c259927fbe4d629a9a94de8080b192bdc17cc13","observation_id":"ed38cf0f-a209-4507-aafa-07872a3f11bf","resolution":{"observed_at":"2026-08-08T13:30:03.547936Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:03.974797Z","title":"Poison frogs! targeted clean-label poisoning attacks on neural networks","venue":null,"work_id":"3ad66c1c-3dbd-441b-b5b1-60c1c56c0bb9","year":2018},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.551583Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:3b0044c4f3011e2a7e81b2ed5ab67e93393a31d9fd34750f14fe44f7486d4be6","observation_id":"6758f524-e135-4270-ac5c-7f9c74a220b5","resolution":{"observed_at":"2026-08-08T13:30:03.978957Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:03.555621Z","title":"Very deep convolutional networks for large-scale image recognition","venue":null,"work_id":null,"year":2015},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.555621Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:688114797de4fc8e8ca8b5143b8ccdaf6f6eb944cfc7ece476c05da8c34313e4","observation_id":"11763ac9-c440-4b5a-8bc1-dd0abc925ee1","resolution":{"observed_at":"2026-08-08T13:30:03.555621Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:03.955714Z","title":"Deep perturbation learning: enhanc- ing the network performance via image perturbations","venue":null,"work_id":"b7cd50a4-45a3-47a7-b698-0141588b697b","year":2023},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.559163Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:90cace86ad52512fa72e063576ad4f4f609b8db97a2fcc865f49c907c4238dd7","observation_id":"fb033342-5c97-4ce5-81eb-88e5a17dbeb5","resolution":{"observed_at":"2026-08-08T13:30:03.959917Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:03.943905Z","title":"The german traffic sign recognition benchmark: a multi-class classification competition","venue":null,"work_id":"b2454ced-e52c-469d-946a-8596430a7d49","year":2011},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.562683Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:a5803a7b3104da010577428021735da58ddad3bc96d18c3a5d472f296c7f2ad4","observation_id":"c6bf342a-0984-4d6f-a7f8-0f5cf839f096","resolution":{"observed_at":"2026-08-08T13:30:03.948024Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:03.931831Z","title":"Mrtrix3: A fast, flexible and open software framework for medical image processing and visualisation","venue":null,"work_id":"a5d01b3d-5ecf-40b8-ae2e-d7384323a201","year":2019},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.566387Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:58f99037348e3ff3d47802e08414e5c7b923502d8d4543285a410e9295f77250","observation_id":"bfe19ca4-db12-4b0b-a473-39e253514113","resolution":{"observed_at":"2026-08-08T13:30:03.936027Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:03.919587Z","title":"Neural cleanse: Identifying and mitigating backdoor attacks in neural networks","venue":null,"work_id":"2a1a00dc-f7fa-4141-bbe8-f0c4de38c0a8","year":2019},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.570626Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:6bc5d5c430f8054db9aeb51615504e1ab60d244eab79f8b8d30a63feb1a776dd","observation_id":"07fa6fc9-0cdf-40f0-8648-d5900054e0cb","resolution":{"observed_at":"2026-08-08T13:30:03.924077Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:03.907557Z","title":"Shared adversarial unlearn- ing: Backdoor mitigation by unlearning shared adversarial examples","venue":null,"work_id":"96ba63da-fd5b-47b8-b5b8-b07d4fcc9fc2","year":2023},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.574359Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:a3d139fce6f16f8b031cb5fd16bec82a7d0b67e29e8f0559dadcd1247c2fe4dd","observation_id":"6839fdd8-e683-4e55-8e4e-63bbf009ce2b","resolution":{"observed_at":"2026-08-08T13:30:03.912017Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2411.09585","last_updated":"2025-02-11T03:32:37Z","snapshot_observed_at":"2026-07-06T19:50:29.247307Z","submitted_at":"2024-11-14T16:54:06Z","title":"Backdoor Mitigation by Distance-Driven Detoxification","version":2},"cited_work":{"arxiv_id":"2411.09585","doi":null,"metadata_source":"pith","pith_arxiv_id":"2411.09585","snapshot_observed_at":"2026-08-08T13:30:03.676113Z","title":"Backdoor Mitigation by Distance-Driven Detoxification","venue":"cs.CR","work_id":"e106499a-8f75-4e7c-bad1-9416a8566996","year":2024},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.582021Z"},"links":{"cited_paper":"/paper/2411.09585","citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:968ab0378a2778292cfc0c9254d0179fa42c879059cc65451c79a13a12c62c90","observation_id":"d342351c-4a0c-46f1-b0ef-967d83f5396b","resolution":{"observed_at":"2026-08-08T13:30:03.680953Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:03.895061Z","title":"Backdoorbench: A comprehensive benchmark of backdoor learning","venue":null,"work_id":"a5d94126-41b8-4007-8305-f526611bb78d","year":2022},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.585807Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:68c6b6b66c04edb0df3f9e179c73e4e511addaea4969809554f71851b03843a7","observation_id":"cc71d9c8-7251-4c7d-935b-e8536b4e0b85","resolution":{"observed_at":"2026-08-08T13:30:03.899616Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:03.881610Z","title":"Defenses in adversarial machine learning: A survey","venue":null,"work_id":"f099d68e-0fd8-4a02-8023-d5349e5cd40b","year":2023},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.589354Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:6534ccf8448efb5f9f9d2b7e9586793b79faccfb1a04bbf55c03256a5f922bfd","observation_id":"308b2868-35f2-44ca-b2b9-144311a148aa","resolution":{"observed_at":"2026-08-08T13:30:03.886477Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:03.870221Z","title":"Backdoorbench: A comprehensive benchmark and analysis of backdoor learning","venue":null,"work_id":"808315ca-6b54-4b3b-87ec-47f9e2255f5a","year":2024},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.592656Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:fb50f86230e4a8197622f580faef674e5935852cf398d5410cac7863b5fb1b8d","observation_id":"9079c694-02c2-4b7c-95e0-522b236a4ee2","resolution":{"observed_at":"2026-08-08T13:30:03.874038Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:03.858994Z","title":"Adversarial neuron pruning purifies backdoored deep models","venue":null,"work_id":"10ac9f36-a549-4654-b0aa-f78f5403a36c","year":2021},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.596001Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:6bbbad20921c7b79b330faa8e3c64a8fca0649b49fc09d5d6a470fbf71c91f8a","observation_id":"70f40f96-c42c-4321-8a2a-5e203d54030c","resolution":{"observed_at":"2026-08-08T13:30:03.862802Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:03.847555Z","title":"Spatially transformed adversarial examples","venue":null,"work_id":"fe31ca74-c69c-48cc-9aa4-a3f6d1ffc254","year":2018},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.599465Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:e5da4abfea3991dbb5c7b0470cbf9a4b183654a9708a04ac5d43f5deb9f6fa01","observation_id":"a744f811-0457-4507-9d36-bcf4b904d6c7","resolution":{"observed_at":"2026-08-08T13:30:03.851707Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:03.834911Z","title":"Morley Mao, and Ruoxi Jia","venue":null,"work_id":"696ea3ae-1365-4043-a319-b5b6460d7153","year":2021},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.602720Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:8e3eee43a9435426b23792c8428621ddd56047a9ec1b64bdef320db369e638df","observation_id":"8e158f87-36dc-47a1-a107-af261f436269","resolution":{"observed_at":"2026-08-08T13:30:03.839214Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:03.822296Z","title":"Adversarial unlearning of backdoors via implicit hypergradient","venue":null,"work_id":"22831425-c4eb-4297-9adb-52fd27875a64","year":2022},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.606273Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:63ccd90b7cd94fed25ca8141cd97c7bb2ffe68bf5b810d7f3c0b1d7e3989a0da","observation_id":"6478dc97-5aee-480e-8130-919607e0eed6","resolution":{"observed_at":"2026-08-08T13:30:03.826589Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2210.06516","last_updated":"2023-05-31T17:58:32Z","snapshot_observed_at":"2026-07-06T14:04:25.335073Z","submitted_at":"2022-10-12T18:18:21Z","title":"How to Sift Out a Clean Data Subset in the Presence of Data Poisoning?","version":2},"cited_work":{"arxiv_id":"2210.06516","doi":null,"metadata_source":"pith","pith_arxiv_id":"2210.06516","snapshot_observed_at":"2026-08-08T13:30:03.655775Z","title":"How to Sift Out a Clean Data Subset in the Presence of Data Poisoning?","venue":"cs.CR","work_id":"5aaa97f9-1266-4d96-92b9-db2a643a743e","year":2022},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.609539Z"},"links":{"cited_paper":"/paper/2210.06516","citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:6e8b5d0cf2dfdaff169fb7e432820cce4e61866dfa73939b4c6834c93863628f","observation_id":"bd6ecc63-9417-4b1d-9dd7-b0a6b11f1700","resolution":{"observed_at":"2026-08-08T13:30:03.662611Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:03.809876Z","title":"Data-free backdoor removal based on channel lipschitzness","venue":null,"work_id":"6a07f2f3-5374-498e-bf9f-23436ee4ea28","year":2022},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":44,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.613663Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:ce0d6ba1e9c1ed74fd486e8cb227eeaed3f07aad1fc9566c0421a4f83f0e6a73","observation_id":"f6cafadd-f1bd-418d-9b75-d228edd7ed3c","resolution":{"observed_at":"2026-08-08T13:30:03.814129Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:03.798219Z","title":"Enhancing fine-tuning based backdoor defense with sharpness-aware minimization","venue":null,"work_id":"97cd7d60-31fb-4452-a40b-bd78fff770b3","year":2023},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":45,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.617072Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:8d1a023e7560c6d380428514934ba4a7b79e0d019bb53c031028084ab8573ef2","observation_id":"ace98698-b558-4d6f-bb32-13b9ea15a6fc","resolution":{"observed_at":"2026-08-08T13:30:03.802195Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:03.786135Z","title":"Neural polarizer: A lightweight and effective backdoor defense via purifying poisoned features","venue":null,"work_id":"7a1d50e4-e356-44c0-91c7-4da13e900651","year":2023},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":46,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.620448Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:92b0c75ad475d206cb51b65d2649847b985f9cd23f7a38964260fb06f4cfe25a","observation_id":"452e2663-6d1f-4e28-8ef9-42aade042626","resolution":{"observed_at":"2026-08-08T13:30:03.790109Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-08T13:30:03.772739Z","title":"Vdc: Versatile data cleanser for detecting dirty samples via visual-linguistic inconsistency","venue":null,"work_id":"76cfc7ba-c6e3-4a4f-9c2c-b52e34e18915","year":2024},"citing_paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification","version":1},"reference_index":47,"source":"pdf_text","source_observed_at":"2026-08-08T13:30:03.623879Z"},"links":{"citing_paper":"/paper/2502.07231"},"observation_digest":"sha256:de690cb5e1c41c6e2eaf2e9d6d2db76730e87e31bd730dca6b7604e347a444ca","observation_id":"793ff486-4d91-47e7-9d83-a434097f2214","resolution":{"observed_at":"2026-08-08T13:30:03.777891Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}}],"paper":{"arxiv_id":"2502.07231","last_updated":"2025-02-11T03:46:35Z","latest_version":1,"primary_category":"cs.CR","snapshot_observed_at":"2026-08-08T13:22:19.556282Z","submitted_at":"2025-02-11T03:46:35Z","title":"Revisiting the Auxiliary Data in Backdoor Purification"},"reference_resolution":{"displayed":46,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":10,"verified_exact":4,"verified_fuzzy":32},"total_outbound_references":46},"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":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"thesis":"As of 9 August 2026, this Paper Citation Record lists 46 of 46 outbound references and 0 inbound Pith citation observations for arXiv:2502.07231."}