{"as_of":"2026-08-13T18:50:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:e29a5326c234458fe29b8df49e48bb06b7b2b14bdfa734165769465bfcec19b2","coverage":[{"denominator":40,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":40,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-10T22:10:36.611106Z","state":"measured"},{"denominator":41,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":41,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-13T06:32:02.005865+00:00","state":"measured"},{"denominator":1,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":1,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-10T22:10:36.611106Z","state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"pith","source_observed_at":"2026-08-10T22:10:36.661421Z","state":"measured"}],"external_citation_measurements":[],"inbound":[{"citation":{"cited_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"cited_work":{"arxiv_id":"2501.02704","doi":null,"metadata_source":"pith","pith_arxiv_id":"2501.02704","snapshot_observed_at":"2026-08-10T22:10:36.661421Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","venue":"cs.LG","work_id":"701ea67e-4507-490a-9b2f-a9c454c00a84","year":2025},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.611106Z"},"links":{"cited_paper":"/paper/2501.02704","citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:7c64a89b203b661548ba0ac2df21f55a844e315a2fa37e4373d50ee917e37897","observation_id":"ef2e456c-5436-4362-a749-323c664cf9bf","resolution":{"observed_at":"2026-08-10T22:10:36.670211Z","resolver_source":"local_arxiv","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}}],"links":{"evidence":"/evidence","html":"/paper/2501.02704/citation-record","integrity":"/paper/2501.02704/integrity","json":"/paper/2501.02704/citation-record.json","paper":"/paper/2501.02704"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T22:10:37.498660Z","title":"Attention is All you Need,","venue":null,"work_id":"a37eaa43-8ba6-4def-b5ed-deb3491987bb","year":2017},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.362661Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:2eb73c2c9d8eb0cfd8b1b3475f16dc0620db1535c00ff30beb69bfc4b4392478","observation_id":"36005ded-ced9-4dc3-9ab7-f793d30ad739","resolution":{"observed_at":"2026-08-10T22:10:37.506124Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:36.369500Z","title":"Language Models are Few-Shot Learners,","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.369500Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:424ed76e8ee62d7c757eeb3d73ca6b2899f179057b71f797b77ea945e41e75a1","observation_id":"03f9c932-07d5-4fe0-a85b-5a4a80c7f43e","resolution":{"observed_at":"2026-08-10T22:10:36.369500Z","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":"2026-08-10T22:10:37.461305Z","title":"GPT-4 Technical Report,","venue":null,"work_id":"84982aeb-209e-4538-a4ea-a89669dca945","year":2024},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.375414Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:dd2b7a0171304fd843ceb0e819dcb7b58b09c09d398bc33fb2ad33c1dc93ebe4","observation_id":"e507fcf3-373b-4824-b0e1-e4b34bcf8cc9","resolution":{"observed_at":"2026-08-10T22:10:37.468724Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:37.437482Z","title":"LaMDA: Language Models for Dialog Applica- tions,","venue":null,"work_id":"b4c11e3b-279c-44e3-b0db-b75f69d74322","year":2022},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.382090Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:24c6774b5c92de130c75ad69a5352ecf408f8d6161da3bf95212fdded4b548c7","observation_id":"9d9b6908-4f5b-48bf-8033-0f3461d5fbf2","resolution":{"observed_at":"2026-08-10T22:10:37.447855Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:37.416705Z","title":"PaLM 2 Technical Report,","venue":null,"work_id":"a7689397-b43d-45f9-8560-a92de452eb25","year":2023},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.388500Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:ce749efcd4f7963e185a22fbecb9cde14664d7ba3294bc512592be5feaa1265a","observation_id":"e21858d6-54d0-4de1-89a4-02240150e1ec","resolution":{"observed_at":"2026-08-10T22:10:37.422307Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:37.396244Z","title":"MLaaS: Machine Learning as a Service,","venue":null,"work_id":"c55b1ca4-2374-4165-ad86-3fad5608665b","year":2015},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.394390Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:54027e006120b56f18472a485c8fd0abc2bcc3d7d31a8b29271a8c187da7e0ff","observation_id":"c2616cb0-c2a3-4267-b92e-b33c026e5299","resolution":{"observed_at":"2026-08-10T22:10:37.402910Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:37.377010Z","title":"Embedding Water- marks into Deep Neural Networks,","venue":null,"work_id":"bd608a27-8413-4cd9-b649-6ff11e38ff77","year":2017},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.400315Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:195eaaf71fc737c1706eb986c0e29ef2c015c135dabb084085df6eea2d90373f","observation_id":"274d7a88-106d-4374-8fdb-d44d5db65bd9","resolution":{"observed_at":"2026-08-10T22:10:37.382355Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:37.358695Z","title":"Turning Your Weakness Into a Strength: Watermarking Deep Neural Networks by Backdooring,","venue":null,"work_id":"2fef7df1-5688-4a32-82aa-d2708c17e141","year":2018},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.408093Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:f0187c9709371edbe655973df50e1c20e0afc90630c9892102a470f6809f125f","observation_id":"1ca1c7af-c62e-43c1-b19f-8f6c16af9715","resolution":{"observed_at":"2026-08-10T22:10:37.364368Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:37.337438Z","title":"On the Robustness of Backdoor-based Watermarking in Deep Neural Net- works,","venue":null,"work_id":"b19cfc81-3771-474d-b0f7-dd39cedf0e8e","year":2021},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.413753Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:aaf94e2081913b22f5841782f3d2fce986028fae3a6c823876cabaac1ea1c8f4","observation_id":"26383cec-7260-4216-afab-63156f3ef695","resolution":{"observed_at":"2026-08-10T22:10:37.345970Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:37.319354Z","title":"REFIT: A Unified Watermark Removal Framework For Deep Learning Systems With Limited Data,","venue":null,"work_id":"faa19cad-e5aa-4f2e-aeaa-bc7362ed8a51","year":2021},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.419387Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:3f23550539f9003bb5dce35af6733a1369d481991b5e88941c9bbb1817d20419","observation_id":"29204acd-abc5-4ee2-8136-c44a9ac577ed","resolution":{"observed_at":"2026-08-10T22:10:37.324844Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:37.300190Z","title":"Fine-Pruning: Defending Against Backdooring Attacks on Deep Neural Networks,","venue":null,"work_id":"8e711dcd-67f6-43df-866b-a7dac8695916","year":2018},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.424624Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:377996d572a67e9fdf288f325e11decc982afc65c86efaca8c1a15f79a4b21c7","observation_id":"7b29bf4c-7a02-43ed-af0e-f7c9dcca9649","resolution":{"observed_at":"2026-08-10T22:10:37.306786Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:37.275838Z","title":"A survey of Deep Neural Network watermarking techniques,","venue":null,"work_id":"91495e54-58fc-469d-a64a-5efe55989cc0","year":2021},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.429918Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:e032a3e18d021eeb0b2d593f4ce4ccfdba334da20364e8ecdbc268f58f353520","observation_id":"e0b51e55-2e9b-4e34-8076-06e0e39a8157","resolution":{"observed_at":"2026-08-10T22:10:37.283537Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:37.257891Z","title":"DeepSigns: An End- to-End Watermarking Framework for Ownership Protection of Deep Neural Networks,","venue":null,"work_id":"7a3e3941-5121-4619-a125-8f832b4cd5d1","year":2019},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.435943Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:7cf2955ffffc41850bb6ebe781a262fb276ae86a65d3dd85f94cb1ca13af16bb","observation_id":"903b8dc8-0092-43f8-b163-e5bdb69121ab","resolution":{"observed_at":"2026-08-10T22:10:37.262844Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:37.234734Z","title":"BadNets: Evaluating Backdooring Attacks on Deep Neural Networks,","venue":null,"work_id":"c74870ec-130d-4bd3-8906-6ff55f963115","year":2019},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.444660Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:a4b686d6758fcfca8929b6e9e99802d9b076811a06b7c19d2cd63a0c3be0fd48","observation_id":"9f8e2758-5e0c-4fe3-8de0-733330d65aa9","resolution":{"observed_at":"2026-08-10T22:10:37.243463Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:37.215990Z","title":"Backdoor Learning: A Survey,","venue":null,"work_id":"bbaca64a-be70-4594-a2cb-b8e0c7362b84","year":2024},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.452550Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:fa6d6fad443b2493fe9283ce1973a5034ad83df377b8fd7ebc9369fa510ddab6","observation_id":"d80493aa-43bd-4861-b551-e7fa8db6fb1c","resolution":{"observed_at":"2026-08-10T22:10:37.220777Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:37.197969Z","title":"Protecting Intellectual Property of Deep Neural Networks with Watermarking,","venue":null,"work_id":"077dc93e-f505-4f81-811a-012576f3af67","year":2018},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.458238Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:a56da6269108603f0c6b76d09be5996e069e31dc55a3bb97c418ef8fa00a88e8","observation_id":"60521784-f4cd-4a84-aede-e25148c04538","resolution":{"observed_at":"2026-08-10T22:10:37.203573Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:37.178597Z","title":"Adversarial frontier stitching for remote neural network watermarking,","venue":null,"work_id":"531b14ef-bac2-496d-9268-0c1b1ae8988e","year":2020},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.465141Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:1282b3b912ad85e33a63c90e51f3222dcb4b48f3cb733c43212732f63a14809a","observation_id":"0d7694b8-afa2-4aef-9668-50297f9393cc","resolution":{"observed_at":"2026-08-10T22:10:37.185330Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:37.161319Z","title":"ROWBACK: RObust Wa- termarking for neural networks using BACKdoors,","venue":null,"work_id":"30deb509-5049-4ec6-89de-a9cc7f719def","year":2021},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.473873Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:aa4c76ed8303c5d7889c01de656dbbae9bd9272e8100cefafadeca3074d9c2a1","observation_id":"301d25ad-0acf-4cca-ac91-6b36cf515ba4","resolution":{"observed_at":"2026-08-10T22:10:37.167053Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:37.142671Z","title":"Explaining and Harnessing Adversarial Examples,","venue":null,"work_id":"9ef29b9c-f4d8-48fd-b78b-0ce1bc7acfa7","year":2015},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.479410Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:f22bf2879bbcd15662cfd8fdc91a28464dd8ccf9ff62801afabbaee8b042d212","observation_id":"67629d85-4fc2-44ef-877c-77c81bae6e4e","resolution":{"observed_at":"2026-08-10T22:10:37.148020Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:36.486469Z","title":"Certified neural network watermarks with randomized smoothing,","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.486469Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:da6d9c4db6d2c1645605e79e5a37f0fc001438275c2ac7680889a169fc595b07","observation_id":"ca1f3d00-1553-4433-beec-c6f30968a825","resolution":{"observed_at":"2026-08-10T22:10:36.486469Z","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":"2026-08-10T22:10:37.110891Z","title":"Dimension-independent Certified Neural Network Watermarks via Mollifier Smoothing,","venue":null,"work_id":"78df98df-df17-4330-9897-ac5b2e970736","year":2023},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.493301Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:088caf302aecac15b64d84566507883daf10d251835dfc251a2f7945640d487e","observation_id":"45d2fdb6-9678-432e-ae77-c1f5039c1124","resolution":{"observed_at":"2026-08-10T22:10:37.117046Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:37.092309Z","title":"Entangled Watermarks as a Defense against Model Extraction,","venue":null,"work_id":"8dc65621-15a2-4ac3-bf7e-44166e82e2a7","year":2021},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.500553Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:f36b71ddc29476e63af7468ffda5a0f244e37d26a1242ebd01f687efa2d0e5c8","observation_id":"f9902528-f84e-4364-a90e-81937f97653d","resolution":{"observed_at":"2026-08-10T22:10:37.098275Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:37.073933Z","title":"Towards Robust Model Watermark via Reducing Parametric Vulnerability,","venue":null,"work_id":"8f541a5d-f8d7-4be0-9036-735b08b9bff4","year":2023},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.505862Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:a7a74207b87f904f857bf414ab6600d92b87289ed70eac9324c2f1065fcc2187","observation_id":"b0b93197-a432-4355-9888-50854ee33d6e","resolution":{"observed_at":"2026-08-10T22:10:37.079695Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:37.055105Z","title":"Free Fine-tuning: A Plug-and-Play Watermarking Scheme for Deep Neural Networks,","venue":null,"work_id":"0c73e5ff-f082-45d2-922f-cc730ba2198b","year":2023},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.511306Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:b56ffd409339bd239a5626fcd8a21b344c45eccd2db9b5140f29b7304787dc6e","observation_id":"cfa41377-ddaa-46ce-ba78-1c6146aafed1","resolution":{"observed_at":"2026-08-10T22:10:37.061303Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:37.031571Z","title":"Cosine Model Watermarking against Ensemble Distillation,","venue":null,"work_id":"f5d35f6e-919a-4232-8c51-727c635bfc63","year":2022},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.517281Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:c3d8ac05582330ff0bf33c081255ac9a29049046d340b5ab3b0f022fa47b83d7","observation_id":"7893d9cf-5f04-49db-a14d-6342387032f4","resolution":{"observed_at":"2026-08-10T22:10:37.040388Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:37.005393Z","title":"Untargeted Backdoor Watermark: Towards Harmless and Stealthy Dataset Copy- right Protection,","venue":null,"work_id":"e3ec3791-f924-4f72-b127-fca7cdf9e04d","year":2022},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.522340Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:63f5d1e965080139d1f62ba6e4ef48dfcdca85bc2140f8fa7c6db778b7991a71","observation_id":"8ee05e90-751a-48bb-a2f1-c45c444824e0","resolution":{"observed_at":"2026-08-10T22:10:37.013220Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:36.984413Z","title":"Unambiguous and High- Fidelity Backdoor Watermarking for Deep Neural Networks,","venue":null,"work_id":"2b40a90e-a7d6-4113-81dc-2ad4696713c4","year":2024},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.530628Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:aaceaefa1ca75611c773c5ee3962341e3d0918abff8ef4df93902cc009ab7dee","observation_id":"84b2ea9e-b5f4-4ec1-a247-cef03ba540c1","resolution":{"observed_at":"2026-08-10T22:10:36.990692Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:36.962410Z","title":"Watermarking Deep Neural Networks in Image Processing,","venue":null,"work_id":"92744f9b-3be6-49d4-9bd2-6be124691ca7","year":2021},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.536017Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:887c5d34d33209363b839e950fb00b3dbd80dcb573134ead42027aba3f0850be","observation_id":"1b9a0a46-d601-49dc-829e-b738675fc5b0","resolution":{"observed_at":"2026-08-10T22:10:36.968414Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:36.926673Z","title":"A Black-Box Watermarking Modulation for Object Detection Models,","venue":null,"work_id":"ae033e24-54ca-4117-bab4-a80859fe1bbf","year":2024},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.541972Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:6043d0849d921cf77f412438eb4a437367dd0098f49a59b28eeeb94106c2c0f4","observation_id":"f14dd3e9-0f24-412b-b938-aac4db1386c1","resolution":{"observed_at":"2026-08-10T22:10:36.935294Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:36.898704Z","title":"Catastrophic Interference in Connec- tionist Networks: The Sequential Learning Problem,","venue":null,"work_id":"529f18ad-07f0-44f1-8bed-67c78859b2cc","year":1989},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.547700Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:4668309fd2876007b3735bf1588c2b1fbefc07ff1596839ff4f3ead63d4e430d","observation_id":"9d299aa9-e6e6-4106-8268-934745e817fc","resolution":{"observed_at":"2026-08-10T22:10:36.908320Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:36.871564Z","title":"Compete to Compute,","venue":null,"work_id":"79e2dab4-3248-4952-b766-87eb9d9ebf50","year":2013},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.552981Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:9046d2b97d17fe72d310edeba3a4f4ec75e183de334998216faf66ee336f99d5","observation_id":"7c6a4b10-35e9-43f6-89b9-64a632c2d2b7","resolution":{"observed_at":"2026-08-10T22:10:36.880062Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:36.851040Z","title":"Overcoming catastrophic forgetting in neural networks,","venue":null,"work_id":"108d11d9-73ac-4109-9408-afdd20b0bb31","year":2017},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.559345Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:3aaea7e2d60c67b265d6c965b2bff58b90b939b9d45eea1aeec7a29632682711","observation_id":"1f9f2c22-01a6-4883-9bdb-bf5cddea1f45","resolution":{"observed_at":"2026-08-10T22:10:36.857915Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:36.827563Z","title":"Ensemble Learning in Fixed Ex- pansion Layer Networks for Mitigating Catastrophic Forgetting,","venue":null,"work_id":"2e7d5154-f1af-4a22-bff7-14f7b77fa8da","year":2013},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.567246Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:f670217ae6948e6d870da7d62868e1ec6a6bdc41aa015b4e3abea83f230b6c72","observation_id":"0adac29f-2bf9-4a9b-8aa2-645ded8713a7","resolution":{"observed_at":"2026-08-10T22:10:36.835911Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:36.796098Z","title":"Measuring catastrophic forgetting in neural networks,","venue":null,"work_id":"b036ab20-7333-4800-bc68-a79a10f04945","year":2018},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.574267Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:fcc180255f746d58f6af4f424785c53de9e9512f99e7ab9ebf690e72ee303b93","observation_id":"9b968c9e-76a5-4ddf-bde7-09df024cf69a","resolution":{"observed_at":"2026-08-10T22:10:36.810521Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-10T22:10:36.579703Z","title":"Learning multiple layers of features from tiny images,","venue":null,"work_id":null,"year":2009},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.579703Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:cab918b341a30fb6cad9b487f91d2c5a37d3ab0fd75ef1cd73196a0f87e454c2","observation_id":"306f98b3-7edd-4f6b-b08d-24892d886966","resolution":{"observed_at":"2026-08-10T22:10:36.579703Z","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":"2026-08-10T22:10:36.585689Z","title":"Reading digits in natural images with unsupervised feature learning,","venue":null,"work_id":null,"year":2011},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.585689Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:a58b66278257fc23533be10b115d903ce9f25f1fd5e2cd14e12796bd38d2a840","observation_id":"8dfc19ac-a951-4983-a2d4-7be2f4b49f01","resolution":{"observed_at":"2026-08-10T22:10:36.585689Z","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":"2026-08-10T22:10:36.592169Z","title":"Deep Residual Learning for Image Recognition,","venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.592169Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:28ab5fcb2e09be760071f15ab6eb3dd4b14fbec3cfdbe16f17c57d4ca0f2c759","observation_id":"fee48e05-9d3a-45f7-9a49-e6761ec0e296","resolution":{"observed_at":"2026-08-10T22:10:36.592169Z","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":"2026-08-10T22:10:36.598133Z","title":"An Image is Worth 16x16 Words: Trans- formers for Image Recognition at Scale,","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.598133Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:6d32ab386cd719fbcb6ca323e0879050b47b0d5e105a6f51a668294d3206b2a3","observation_id":"091fdb3e-8871-4eff-be17-248c44bbe4f0","resolution":{"observed_at":"2026-08-10T22:10:36.598133Z","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":"2026-08-10T22:10:36.702065Z","title":"SoK: How Robust is Image Classification Deep Neural Network Watermarking?","venue":null,"work_id":"bf8feb3d-7976-4742-9d50-4f7b25fb79ed","year":2022},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.604255Z"},"links":{"citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:123aeb032488a809ca53a6a5cc30d4ef563c38da387f96ee0b7b141032baec3d","observation_id":"244924cd-1378-4d4d-b522-f671961044fb","resolution":{"observed_at":"2026-08-10T22:10:36.707928Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"cited_work":{"arxiv_id":"2501.02704","doi":null,"metadata_source":"pith","pith_arxiv_id":"2501.02704","snapshot_observed_at":"2026-08-10T22:10:36.661421Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","venue":"cs.LG","work_id":"701ea67e-4507-490a-9b2f-a9c454c00a84","year":2025},"citing_paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation","version":3},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-08-10T22:10:36.611106Z"},"links":{"cited_paper":"/paper/2501.02704","citing_paper":"/paper/2501.02704"},"observation_digest":"sha256:7c64a89b203b661548ba0ac2df21f55a844e315a2fa37e4373d50ee917e37897","observation_id":"ef2e456c-5436-4362-a749-323c664cf9bf","resolution":{"observed_at":"2026-08-10T22:10:36.670211Z","resolver_source":"local_arxiv","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}}],"paper":{"arxiv_id":"2501.02704","last_updated":"2025-05-09T06:46:46Z","latest_version":3,"primary_category":"cs.LG","snapshot_observed_at":"2026-08-12T15:20:17.490673Z","submitted_at":"2025-01-06T01:15:35Z","title":"Persistence of Backdoor-based Watermarks for Neural Networks: A Comprehensive Evaluation"},"reference_resolution":{"displayed":40,"state_counts":{"malformed_identifier":0,"metadata_mismatch":1,"parse_uncertain":0,"unresolved":6,"verified_exact":0,"verified_fuzzy":33},"total_outbound_references":40},"refusal":"A citation records a reference. It does not transfer a finding from one paper to another.","schema":"pith.paper-citation-record.v1","standing_sources":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"thesis":"As of 13 August 2026, this Paper Citation Record lists 40 of 40 outbound references and 1 inbound Pith citation observation for arXiv:2501.02704."}