{"as_of":"2026-08-16T04:40:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:907e9b10a2f216a6236df31fca03939a0600caa5822484d6611bfade4e528bbb","coverage":[{"denominator":70,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":70,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-12T10:13:31.810917Z","state":"measured"},{"denominator":70,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":70,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-15T06:32:42.880941+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/2411.19479/citation-record","integrity":"/paper/2411.19479/integrity","json":"/paper/2411.19479/citation-record.json","paper":"/paper/2411.19479"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T10:13:34.799610Z","title":"Physgan: Generating physical-world-resilient adversarial examples for autonomous driving","venue":null,"work_id":"a2a0645e-e06c-4316-9347-af00997ae87b","year":2020},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:30.735476Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:d12d4b4340bd5181b24ecca8a65d2dcc8878e4c50fe4d6553f5a703bf6dd30db","observation_id":"d81b2737-7d86-4d0e-af46-431c9dd69b50","resolution":{"observed_at":"2026-08-12T10:13:34.804105Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:34.784032Z","title":"A survey of deep learning techniques for au- tonomous driving.J","venue":null,"work_id":"0831fbcd-1c64-4988-a046-5164ba1cf3ed","year":2020},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:30.832398Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:4d3229b00b2edd376d75d3c03a6ce00a675d5e6a2d6e5183e990bdf53bb990c2","observation_id":"e6c45984-efb2-4bab-8354-7f7a3f7da316","resolution":{"observed_at":"2026-08-12T10:13:34.792427Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:34.704718Z","title":"Mutual component analysis for heterogeneous face recog- nition.ACM Trans","venue":null,"work_id":"88e9422b-ceaa-4573-ae79-40ec925dcf05","year":2016},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:30.841673Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:6ac81664f566206080d87443b1615272de05a382305c44dc05608335b1e29a12","observation_id":"dfe671ff-5d59-4c21-bff5-21e135b6fc7d","resolution":{"observed_at":"2026-08-12T10:13:34.738066Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:34.684456Z","title":"Com- mon feature discriminant analysis for matching infrared face images to optical face images.IEEE Trans","venue":null,"work_id":"f9211334-caa7-4abf-abfe-60925888ea23","year":2014},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:30.845217Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:1c40e116dc703d903d123141f61203d54aa4078cdfa96fe35eb5d6b2aaececfb","observation_id":"b2acb4b4-edad-4f9b-b677-f184f66d660d","resolution":{"observed_at":"2026-08-12T10:13:34.689781Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:34.625769Z","title":"Detecting and corrupting convolution-based unlearnable examples","venue":null,"work_id":"149fee57-f3fd-4634-b544-b134c6dba53a","year":2023},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:30.848805Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:3473cba4b5016da025a182d541853f5581fa833c6bbbf75cd3266214538a4ad4","observation_id":"68805275-57f8-4d4d-ba6c-97e65212689e","resolution":{"observed_at":"2026-08-12T10:13:34.669798Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:34.596904Z","title":"Badrobot: Manipulating embodied llms in the physical world","venue":null,"work_id":"4b0f54fa-47f8-4043-b040-b883086fa034","year":2024},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:30.851915Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:2915ecbb6b3b79c2b1d2aaf7472f509a1bf5716ffcc121843524f453f757c93a","observation_id":"ea6b32dc-89ce-4b4d-a054-384086559cfc","resolution":{"observed_at":"2026-08-12T10:13:34.601235Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:34.584074Z","title":"Decaf: Data distribution decompose attack against federated learning","venue":null,"work_id":"4832b850-f671-4d10-8905-bec2dff2ece2","year":2025},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:30.855467Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:c4531d2fbf1c748fb515876857e1c90a3853bb58a9c74001d372e0a1f09b11be","observation_id":"00a13fc2-e729-4796-a819-5102b1179f9d","resolution":{"observed_at":"2026-08-12T10:13:34.588729Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:34.512406Z","title":"Improving generaliza- tion of universal adversarial perturbation via dynamic maximin optimization","venue":null,"work_id":"6fc225df-eab5-41c4-8bae-fc7de3a0bfd8","year":2025},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:30.858644Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:4a01bcf613c61de7bc71d3947ec9632c637e9bfcce8f2e63c4daa283cf170b58","observation_id":"0b8f6000-a936-48e1-9ab3-ae479fe685e5","resolution":{"observed_at":"2026-08-12T10:13:34.555772Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:34.432280Z","title":"Pb-uap: Hybride universal adversarial attack for image segmentation","venue":null,"work_id":"6480e7d7-4d71-413f-b9f0-eb6cd71dc247","year":2025},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:30.862791Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:94890cabc2deb0a31d298f438079ba7d68f512577dd8a5ce8662dfbc4c3f014f","observation_id":"0fbd17f6-4a86-4a3c-b76e-910752285ddb","resolution":{"observed_at":"2026-08-12T10:13:34.436009Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:34.419843Z","title":"Bad- Nets: Identifying Vulnerabilities in the Machine Learning Model Supply Chain.IEEE Access, 2017","venue":null,"work_id":"2fc79155-305e-4ccd-865d-3395b5389bd0","year":2017},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:30.866564Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:f5864a376862eaf352a8005b6c82ec8ab0447ebc8225ab7f92a8ffab747fc6e6","observation_id":"1d8a7bb2-679d-4508-80c0-92340092a7c6","resolution":{"observed_at":"2026-08-12T10:13:34.423761Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:34.406322Z","title":"Backdoor Learning: A Survey.IEEE Trans","venue":null,"work_id":"24627f2a-ae35-4516-b62c-3982e5120802","year":2022},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:30.870211Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:cdb0ec1291b91621f89afe49eff478c8e1c3989bf1d76cee08e89dea8092fd16","observation_id":"8cad499b-7477-490e-9b51-49403805bc88","resolution":{"observed_at":"2026-08-12T10:13:34.410279Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:34.358277Z","title":"Pointncbw: Towards dataset ownership verification for point clouds via negative clean-label backdoor watermark.IEEE Trans","venue":null,"work_id":"72324aae-f43d-4b81-a5fb-248bddef7035","year":2024},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:30.874647Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:0396753452d4ff91d77f41ccdb8d0e9be19051ec27775b906500249052099f8e","observation_id":"978b7f2f-dadc-4986-8a01-049512d3a858","resolution":{"observed_at":"2026-08-12T10:13:34.396599Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:34.252893Z","title":"Backdoor attack with sparse and invisible trigger.IEEE Trans","venue":null,"work_id":"920198a7-4de3-42e3-b320-b255a4e2200d","year":2024},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:30.878603Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:9a73536ed24dad8f7d8132a0a465ccb7026404182dbe1a5ffe4f1a6246efe9cb","observation_id":"db5fc1f8-123e-4180-a1a0-de0e74997c09","resolution":{"observed_at":"2026-08-12T10:13:34.306025Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:34.168532Z","title":"Just a little human intelligence feedback! unsupervised learning assisted supervised learning data poisoning based backdoor removal","venue":null,"work_id":"633e5b5f-44cd-4754-8818-226966f8dc3d","year":2025},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:30.882043Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:f1ac0625cd07e60eb96c11832148d6410d626bfaecce8cb02e85387a8937290e","observation_id":"0883453a-cbf2-45ce-9057-a60e7fe7fdcf","resolution":{"observed_at":"2026-08-12T10:13:34.213676Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:34.129861Z","title":"Towards a proactive ML approach for detecting backdoor poison samples","venue":null,"work_id":"037f9374-eba6-4a23-9ec4-7eafc50404e4","year":2023},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:30.939520Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:b6f459e66bc6f774a8584e582f9a8a06885d080a9b1c9326b0020e04f44ad049","observation_id":"8b966ec4-fc1d-4e28-ac19-b518db05155d","resolution":{"observed_at":"2026-08-12T10:13:34.134577Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:34.112110Z","title":"Reverse backdoor distillation: Towards online backdoor attack detection for deep neural network models.IEEE Trans","venue":null,"work_id":"a75b510b-2f30-4e32-8ef2-106bbb45b4d5","year":2024},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.024803Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:cc7f83f0978f04126a40613181892fe8e22a9279bd9036a977022a98edf49dbf","observation_id":"ab9d5604-7f4b-4288-b2d4-99599068cfa9","resolution":{"observed_at":"2026-08-12T10:13:34.119124Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:34.084175Z","title":"IBD-PSC: Input-level backdoor detec- tion via parameter-oriented scaling consistency","venue":null,"work_id":"995f7d19-ba23-4f92-a4b8-2230c12f2899","year":2024},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.062589Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:c707c705ddb4e4f7323f8a65061fec4f197822e751e4adf066f55fac2f639ff9","observation_id":"89c0890f-5726-4b1a-9aa0-df8dcef23a68","resolution":{"observed_at":"2026-08-12T10:13:34.103168Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:31.067657Z","title":"Backdoor defense via decoupling the training process","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.067657Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:5bf2500d1f3a25f8c13b32f177bba764fe37cf3c356d67b8abae235225727bad","observation_id":"a7c7040d-3e65-4b2b-b1b3-3f62e9d488f7","resolution":{"observed_at":"2026-08-12T10:13:31.067657Z","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":"2026-08-12T10:13:33.962756Z","title":"Backdoor defense via adaptively splitting poisoned dataset","venue":null,"work_id":"155c37d6-38ee-4bf8-a778-c47705321c5e","year":2023},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.071468Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:f487a65fb90fcec08d6234f00a1d7792a775642ad32cd655d054fbde7cabc675","observation_id":"23db8e27-2b58-4970-a0ce-f5e901ee5de3","resolution":{"observed_at":"2026-08-12T10:13:33.997816Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:33.951230Z","title":"Setting the Trap: Capturing and Defeating Backdoor Threats in PLMs through Honeypots","venue":null,"work_id":"6baad9c1-675b-4061-8390-16f4bf8cdf9f","year":2023},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.075293Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:b611b5b165a297b90d0d6040ce89c8ac3e75078b52f301f883af40766e8aa525","observation_id":"7a5b2db7-ec27-4e5f-a909-74e833dd54b7","resolution":{"observed_at":"2026-08-12T10:13:33.955462Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:33.938771Z","title":"Neural Cleanse: Identifying and Mitigating Backdoor Attacks in Neural Net- works","venue":null,"work_id":"8282638e-fc25-4e84-875c-1c3f4aa37747","year":2019},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.078590Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:1961ab519c6382df8826c55b60fbda52047e84c2bb835f4a8ceff361a1ea029f","observation_id":"e4ac43df-aa81-405b-8df6-5a18d2b25c32","resolution":{"observed_at":"2026-08-12T10:13:33.943370Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:33.927047Z","title":"Umd: Unsupervised model detection for x2x backdoor attacks","venue":null,"work_id":"e3e09a18-2c78-43f6-a535-cea70f73d9c8","year":2023},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.084166Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:c14e0b5532c00f765835821c1db224c86dcb01512a397fc00943fff9536f3b2d","observation_id":"0fed7950-41c6-4846-9b89-5cc94b5a6300","resolution":{"observed_at":"2026-08-12T10:13:33.931239Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:33.902734Z","title":"MM-BD: Post-Training Detection of Backdoor Attacks with Arbitrary Backdoor Pattern Types Using a Maximum Margin Statistic","venue":null,"work_id":"5b97e3f9-4a71-4bd3-83cf-8900fdfee405","year":2024},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.087419Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:674b13324a74a72aaccc33a3cb2708e08276e95abf0934d58f540b4f62cede1a","observation_id":"5dad7f61-f3d1-45c4-ad70-a1071cc330d7","resolution":{"observed_at":"2026-08-12T10:13:33.919155Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:33.819853Z","title":"SCALE-UP: An efficient black-box input-level backdoor detection via analyzing scaled prediction consistency","venue":null,"work_id":"f7f29e67-f252-4493-b472-fbac3e78f997","year":2023},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.090864Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:8662c3c2b67e5a5b7a05844bbdff62479a2149fdd49abb31236d801d4a0c7281","observation_id":"e839ac97-e1ee-4603-a9fc-01f8da01c23e","resolution":{"observed_at":"2026-08-12T10:13:33.857616Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:33.808057Z","title":"Backdoor secrets unveiled: Identifying backdoor data with optimized scaled prediction consistency","venue":null,"work_id":"acc9730e-3dab-4720-8528-7116c15eef89","year":2024},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.094766Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:8badef28b95257856aca0cdc9db6fe5c1e9e280340c2a8ed0802c93eb7465630","observation_id":"5132ebae-646b-4f9b-baec-3736ea8e3207","resolution":{"observed_at":"2026-08-12T10:13:33.811977Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:33.796089Z","title":"Fine- pruning: Defending against backdooring attacks on deep neural networks","venue":null,"work_id":"dacddf49-8595-4380-846b-6f9f18280383","year":2018},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.098764Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:8cd3ab72262214dde0519988158ec4ba910203fafa91fd242fd45e2e956e4240","observation_id":"f1b340bb-296f-4c25-a7b0-5161cfbc274f","resolution":{"observed_at":"2026-08-12T10:13:33.800505Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:33.777334Z","title":"Adversarial Unlearning of Backdoors via Implicit Hypergradient","venue":null,"work_id":"1c5396d6-10ba-4e12-bf59-035b2f7639f7","year":2022},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.102227Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:c462ab680805a3380dc10124f27cac2be9080abfb40e50496e35aac606580252","observation_id":"d87c31df-ce54-4647-9c3f-c11138d6bc0a","resolution":{"observed_at":"2026-08-12T10:13:33.787571Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:31.106048Z","title":"Towards reliable and efficient backdoor trigger inversion via decoupling benign features","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.106048Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:f24859c978a8c123a1bd8492f062b201c53bc8169723148523d95d27d7549b1d","observation_id":"ea31ca73-aa73-4592-830e-b00821eb41eb","resolution":{"observed_at":"2026-08-12T10:13:31.106048Z","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":"2026-08-12T10:13:33.703403Z","title":"Anti-Backdoor Learning: Training Clean Models on Poisoned Data","venue":null,"work_id":"713030a6-d39c-4340-a3db-9afc1c030b17","year":2021},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.155588Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:ba0428ec2be9bbb8926e7c66d16f23c8587b99dfae28cbdea448b6b9425db6f6","observation_id":"374d4d93-eff3-459a-bb79-3eea64fe8bef","resolution":{"observed_at":"2026-08-12T10:13:33.742011Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:33.655693Z","title":"Backdoor defense via deconfounded representation learning","venue":null,"work_id":"079fb40a-2b6c-42d7-b743-4e3dc555301d","year":2023},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.248346Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:4e5418f3e128b051ece7dbb895f98c89fca941f5d9a2e51bf980b14fd22fe6fe","observation_id":"9c78c96e-0f56-4235-84ad-d0d4953c0298","resolution":{"observed_at":"2026-08-12T10:13:33.659486Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:33.644719Z","title":"Sen- tiNet: Detecting Localized Universal Attacks Against Deep Learning Systems","venue":null,"work_id":"1f117945-3371-4eb0-9a1b-508d38c201ce","year":2020},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.258498Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:a428b4d9ebc3f87f940fd832696f1a60b0e34086d0261072b144d1537fae0b05","observation_id":"ef340c87-3eb0-4b0f-8eac-fb33318d4cf8","resolution":{"observed_at":"2026-08-12T10:13:33.649183Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:33.631813Z","title":"Distilling cognitive backdoor patterns within an image","venue":null,"work_id":"98594abe-b4a3-4f39-8cd2-b8ed98d5e092","year":2023},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.264133Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:abf1e9ae5b5532cc7b71bd9c1a122c811fbbd15144eab8b07ceca089a5d51419","observation_id":"5c6b37dc-2d59-4079-a26a-f4e266cf0393","resolution":{"observed_at":"2026-08-12T10:13:33.636077Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:33.592608Z","title":"Can neural network memorization be localized? InICML, 2023","venue":null,"work_id":"c3b95957-be18-4169-8d8a-b3376876de4c","year":2023},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.267675Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:4914afdf63587ab1a7985bffe11974379dc32812811d2c0d2d01ec889d3bfc9a","observation_id":"ab1ea636-6569-409f-b28b-f32548c1a33f","resolution":{"observed_at":"2026-08-12T10:13:33.619973Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2503.04853","last_updated":"2025-03-06T06:00:04Z","snapshot_observed_at":"2026-08-07T17:25:46.864744Z","submitted_at":"2025-03-06T06:00:04Z","title":"From Pixels to Trajectory: Universal Adversarial Example Detection via Temporal Imprints","version":1},"cited_work":{"arxiv_id":"2503.04853","doi":null,"metadata_source":"pith","pith_arxiv_id":"2503.04853","snapshot_observed_at":"2026-08-12T10:13:31.941330Z","title":"From Pixels to Trajectory: Universal Adversarial Example Detection via Temporal Imprints","venue":"cs.CR","work_id":"3b800a76-8dc3-4f6f-b550-32b83fc7d242","year":2025},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.270711Z"},"links":{"cited_paper":"/paper/2503.04853","citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:d2ebf59099c216d7c2a61295c7a1584f1e21b96d67d59634196ec123aa8ec9e6","observation_id":"a1720868-8f60-45ff-88ff-abd20ea5f93d","resolution":{"observed_at":"2026-08-12T10:13:31.998560Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:33.490655Z","title":"Why does little robustness help? a further step towards understanding adversarial transferability","venue":null,"work_id":"dae5bf99-959a-43d1-b86b-1cdc8a4d4cec","year":2024},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.274733Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:44274581a4c3fec1a113bc81d947c9d247032f98a0213391e0ae0f83044f9e9f","observation_id":"6f1ef045-005e-449b-8191-a6de639a8d32","resolution":{"observed_at":"2026-08-12T10:13:33.563133Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:33.316889Z","title":"Towards label-only membership inference attack against pre-trained large language models","venue":null,"work_id":"efb06eb8-9e0a-4715-87fa-c28862339008","year":2025},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.279545Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:2057eac67886fdba28f689813814ad4424c15e2fac39e924a531b576e0d18887","observation_id":"0fef7c1e-f8c9-4c38-9ff1-827e78180e96","resolution":{"observed_at":"2026-08-12T10:13:33.379227Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:33.304567Z","title":"Yes,{One- Bit-Flip}matters! universal{DNN}model inference depletion with runtime code fault injection","venue":null,"work_id":"eb10d976-b177-4795-8c64-9a3fda85158b","year":2024},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.284361Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:0f17b1da134666047cfd12fb3613343f4f93518fcea797ca4321aba01e7c8e70","observation_id":"ef615e06-6beb-4fbb-b4d8-e72e87b66fe1","resolution":{"observed_at":"2026-08-12T10:13:33.309298Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2007.10760","last_updated":"2020-08-02T08:38:25Z","snapshot_observed_at":"2026-08-14T12:42:44.861746Z","submitted_at":"2020-07-21T12:49:12Z","title":"Backdoor Attacks and Countermeasures on Deep Learning: A Comprehensive Review","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2007.10760","snapshot_observed_at":"2026-08-12T10:13:31.347700Z","title":"Back- door attacks and countermeasures on deep learning: A compre- hensive review.arXiv preprint arXiv:2007.10760, 2020","venue":null,"work_id":null,"year":2007},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.347700Z"},"links":{"cited_paper":"/paper/2007.10760","citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:abc9226237ca406ed42c71b349dd68e7d50c7f2ec352b058eca2f82f5819c488","observation_id":"e3da99a3-dd56-4309-adf5-38d27efdc916","resolution":{"observed_at":"2026-08-12T10:13:31.347700Z","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":"2026-08-12T10:13:33.294467Z","title":"WaNet – Imperceptible Warping-based Backdoor Attack","venue":null,"work_id":"e01dac48-a175-41fa-944b-6e9086fa1d93","year":2021},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.451903Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:31ad9950dbe533b8d050bc329a4f20be61bf52f34f52740fa784962317e419a0","observation_id":"0ab53d8f-8e9e-4240-96d4-87c396a8bdc5","resolution":{"observed_at":"2026-08-12T10:13:33.298145Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1912.02771","last_updated":"2019-12-06T23:16:45Z","snapshot_observed_at":"2026-08-10T18:11:31.253134Z","submitted_at":"2019-12-05T18:05:59Z","title":"Label-Consistent Backdoor Attacks","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1912.02771","snapshot_observed_at":"2026-08-12T10:13:31.456740Z","title":"Label-consistent backdoor attacks.arXiv preprint arXiv:1912.02771, 2019","venue":null,"work_id":null,"year":1912},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.456740Z"},"links":{"cited_paper":"/paper/1912.02771","citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:c822cdb93b34b53da1ae7d005be634c6722f2acb944c1395e4f694ad792b243f","observation_id":"00ffa933-b827-4aee-b4d9-3fbcf63428ca","resolution":{"observed_at":"2026-08-12T10:13:31.456740Z","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":"2026-08-12T10:13:33.283451Z","title":"Invisible Backdoor Attack with Sample-Specific Triggers","venue":null,"work_id":"142b0067-b7d6-4e12-864d-6843c3c9f1bb","year":2021},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.463256Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:be67aa1aa842d514f2407207e2931d4b2e31e63e350e4c44ce3d82c127487895","observation_id":"595b0c75-6951-4b71-9606-ad9cc9b1e246","resolution":{"observed_at":"2026-08-12T10:13:33.287532Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:33.271334Z","title":"Watch out! simple horizontal class backdoor can trivially evade defense","venue":null,"work_id":"7b6d74c2-416e-4af8-8bf7-f017f98f7bc6","year":2024},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.466978Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:58ab6b7d4972566d64428dcbc2f3f6114d6e3630d57d50a46f133e66445874c7","observation_id":"837ae483-cac9-4f67-bbfb-285258b21f9b","resolution":{"observed_at":"2026-08-12T10:13:33.275685Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:33.245250Z","title":"Revisiting the Assumption of Latent Separability for Backdoor Defenses","venue":null,"work_id":"6a03ffca-5f67-42c6-8a39-8218c30f528c","year":2023},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.470542Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:f943655ec75cc8fd834cb640a6a562efe125dc4165404d70a91c399afa210d36","observation_id":"c827f207-2b19-4d78-9d63-691df306a28c","resolution":{"observed_at":"2026-08-12T10:13:33.264600Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:33.142882Z","title":"Untargeted backdoor watermark: Towards harmless and stealthy dataset copyright protection","venue":null,"work_id":"832fd2c6-4a24-4bbb-bbbd-0a72442cbebf","year":2022},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":44,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.476581Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:d60dc5fdf983701eb28367f03a5185f545c5ce58ffbeb672bef593519c4da5c9","observation_id":"fada85f3-9287-453e-929f-b5663729fa60","resolution":{"observed_at":"2026-08-12T10:13:33.209615Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:32.959207Z","title":"Untargeted backdoor attack against deep neural networks with imperceptible trigger.IEEE Trans","venue":null,"work_id":"35d08e48-39eb-4723-b470-c707fda7d9a0","year":2023},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":45,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.481143Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:6f600a60fe4617d8ef96380072d3a9af0d7708e78c5169fd617e0fab2242f8b5","observation_id":"e713f631-ebbb-4e0a-8b16-2fb2e0c7fd57","resolution":{"observed_at":"2026-08-12T10:13:33.075833Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:32.816658Z","title":"Try to poison my deep learning data? nowhere to hide your trajectory spectrum! InNDSS, 2025","venue":null,"work_id":"2c8bb893-d43b-4b50-89cf-d09b5c19e598","year":2025},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":46,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.484610Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:124539436b8e1ecef6f45765363c34b4e774d9daf9db6a303164a5e0cd7c37f7","observation_id":"1c89bcc6-7a23-4685-b2a6-7314b686fe06","resolution":{"observed_at":"2026-08-12T10:13:32.864863Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:32.807055Z","title":"Strip: A defence against trojan attacks on deep neural networks","venue":null,"work_id":"3b6fdf51-315b-4f57-9519-6e3bbb51b5cb","year":2019},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":47,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.487685Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:affb3003e536c909c85dad5635755a2236a3330a2d4aecf4fff56a9013b8c548","observation_id":"4b7d977b-cd6e-4aec-b717-1015eeca73c2","resolution":{"observed_at":"2026-08-12T10:13:32.810330Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:32.795906Z","title":"Ntd: Non-transferability enabled deep learning backdoor detection.IEEE Trans","venue":null,"work_id":"a89909c8-7191-495a-aa4b-166241d08e67","year":2023},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":48,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.491326Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:f3f4e9597fe8c99c972c595094ac7dc54aac8750ce7cf5d410e08c78a0bc17d1","observation_id":"0543f87d-2826-42b9-8f44-4bcc516e1600","resolution":{"observed_at":"2026-08-12T10:13:32.799911Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:32.784644Z","title":"Purifying quantization-conditioned backdoors via layer-wise activation correction with distribution approximation","venue":null,"work_id":"326c302f-6718-42a5-bd72-9938273b7cd8","year":2024},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":49,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.496612Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:b31311bf321ea90b0368644eb07aaccc6ac54a7c6a1796a4c5cb715164051001","observation_id":"9f2b03f4-24c4-4a32-a33f-76820030b736","resolution":{"observed_at":"2026-08-12T10:13:32.788805Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:32.771163Z","title":"Detecting backdoor attacks on deep neural networks by activation clustering","venue":null,"work_id":"42611de4-d345-49be-b03a-b54ce641549c","year":2018},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":50,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.500734Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:8b8162953bafd991d463c1fda18d0cbdb7fb5e6b3ec84181abf42f1ccc932813","observation_id":"2323d8a4-ab6b-4c25-bf26-64f9a6ad55ed","resolution":{"observed_at":"2026-08-12T10:13:32.775302Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:32.680609Z","title":"The “beatrix”resurrections: Robust backdoor IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY 15 detection via gram matrices","venue":null,"work_id":"fcc8dfc5-8b81-4442-9c49-d904a857fed8","year":2022},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":51,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.504514Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:59295f97b5b2a0215b62c3193fdec8336a5fdd3cc6e93f238888e96bbc583765","observation_id":"fcd27572-9691-40df-9c65-43742fab30a2","resolution":{"observed_at":"2026-08-12T10:13:32.758634Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:32.466583Z","title":"Grad- cam: Visual explanations from deep networks via gradient-based localization","venue":null,"work_id":"dd363e8b-8dda-4bb6-823c-2042774ac58f","year":2017},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":52,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.528436Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:b1a838d995d520c998b4db89509e91e0419a3f35b1ebf54aca88cb5354a88abe","observation_id":"7b1ea460-f845-4f80-85a8-f6ec5a9713e9","resolution":{"observed_at":"2026-08-12T10:13:32.594181Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:32.427145Z","title":"Adversarial neuron pruning purifies backdoored deep models","venue":null,"work_id":"e5a4d6e5-32b8-4ca0-b5d3-5e34ea0aeca1","year":2021},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":53,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.597136Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:c60e537cf57174c686ba344a9f4706c4af3741c112f4241cd41c144ecf965e94","observation_id":"817a30f7-8de4-48f8-842f-2e531bfaf096","resolution":{"observed_at":"2026-08-12T10:13:32.431430Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:32.415256Z","title":"One-shot neural backdoor erasing via adversarial weight masking","venue":null,"work_id":"88901e06-db4b-43e7-89e4-cb42c1c8e943","year":2022},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":54,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.667960Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:37907442a17fb25ed616d348a23f2fce234e3be7c7427fc1b8eef6a91f92ef7d","observation_id":"7de12a30-e3a7-4157-b889-e402e9796a78","resolution":{"observed_at":"2026-08-12T10:13:32.420001Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:32.402678Z","title":"Neural attention distillation: Erasing backdoor triggers from deep neural networks","venue":null,"work_id":"7a9b4f1d-290c-4a18-a5dd-235771e44c86","year":2021},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":55,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.679671Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:2e3205fcb0f28776cf91677e0fc58cf9d70f5636c2b7381efb0ff73c2ec1b829","observation_id":"5e33075a-ffd1-415d-8d5d-3edc892342e1","resolution":{"observed_at":"2026-08-12T10:13:32.406620Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:32.388555Z","title":"Breaking the false sense of security in backdoor defense through re-activation attack","venue":null,"work_id":"e0d5098f-95dd-490a-b070-cfaf78db21d7","year":2024},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":56,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.691637Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:670239fad612ef91b39f1102f21ed75c2fc1be475398c178efa8550758eee877","observation_id":"3e5c39f7-beed-4d06-94c3-45f948dda046","resolution":{"observed_at":"2026-08-12T10:13:32.393626Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:32.255793Z","title":"Randomized channel shuffling: Minimal- overhead backdoor attack detection without clean datasets","venue":null,"work_id":"1d678db5-f38e-497d-ac38-8f5dfc409dff","year":2022},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":57,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.705058Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:c8191f1cd6757ee83272303c8382e71e2a8d94b58d52af74742038cc61407dcd","observation_id":"088dc1f1-2942-4f6b-bda4-8fc5a6ba3e9f","resolution":{"observed_at":"2026-08-12T10:13:32.349932Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:31.711543Z","title":"Umap: Uniform manifold approximation and projection for dimension reduction.J","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":58,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.711543Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:d401b97a0420fb7e20aa1ebf0657a2fa1120f50bfb3b48eaa85b52f8e1cd7eac","observation_id":"3441a424-8e88-4122-ab76-c150897b7ec2","resolution":{"observed_at":"2026-08-12T10:13:31.711543Z","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":"2026-08-12T10:13:31.714985Z","title":"hdbscan: Hierarchical density based clustering.J","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":59,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.714985Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:07bfffc7029435bab0475e1fc9d11cc83509b93d28d188c05ae2577f44e5288d","observation_id":"34fbdcf0-c98f-4bbc-9932-64e9c646f90f","resolution":{"observed_at":"2026-08-12T10:13:31.714985Z","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":"2026-08-12T10:13:32.155662Z","title":"Learning Multiple Layers of Features from Tiny Images.Tech","venue":null,"work_id":"c6e74fc8-0202-4dad-9384-1de6ee866bad","year":2009},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":60,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.719656Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:12f48a641fdd87a988c4e414049c14c6c383b6afb46a4329d2e53318a64a6dc7","observation_id":"65aae897-3eae-47c1-b55f-0d183b900b83","resolution":{"observed_at":"2026-08-12T10:13:32.179196Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1707.08819","last_updated":"2017-08-23T16:06:20Z","snapshot_observed_at":"2026-08-14T20:44:55.373539Z","submitted_at":"2017-07-27T11:22:22Z","title":"A Downsampled Variant of ImageNet as an Alternative to the CIFAR datasets","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1707.08819","snapshot_observed_at":"2026-08-12T10:13:31.746060Z","title":"A downsampled variant of imagenet as an alternative to the cifar datasets.arXiv preprint arXiv:1707.08819, 2017","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":61,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.746060Z"},"links":{"cited_paper":"/paper/1707.08819","citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:d3f43dbfff4e2d14193e379bdd43712333f227821c860e3ff61d3aef5fe129c6","observation_id":"62fefa54-949e-4083-8a48-7a248b397351","resolution":{"observed_at":"2026-08-12T10:13:31.746060Z","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":"2026-08-12T10:13:32.143563Z","title":"Deep Residual Learning for Image Recognition","venue":null,"work_id":"b186d422-5937-4572-a52b-0129724d68fd","year":2016},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":62,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.750092Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:48951d4b1f7e75e7fb742f94199daf6f62d5f65c6ed068902a832008864c0c98","observation_id":"d4116b3d-e2f9-4559-b83a-fbab5316b151","resolution":{"observed_at":"2026-08-12T10:13:32.147840Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1409.1556","last_updated":"2015-04-10T16:25:04Z","snapshot_observed_at":"2026-08-14T23:20:42.336514Z","submitted_at":"2014-09-04T19:48:04Z","title":"Very Deep Convolutional Networks for Large-Scale Image Recognition","version":6},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1409.1556","snapshot_observed_at":"2026-08-12T10:13:31.754690Z","title":"Very deep convolu- tional networks for large-scale image recognition.arXiv preprint arXiv:1409.1556, 2014","venue":null,"work_id":null,"year":2014},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":63,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.754690Z"},"links":{"cited_paper":"/paper/1409.1556","citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:8d73f59df873d5427413456580e3f7f58f773e712d351fbfcdc5e8e1768688ab","observation_id":"56ce3f61-4999-4fa5-b6ef-d5b4bc394d0a","resolution":{"observed_at":"2026-08-12T10:13:31.754690Z","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":"2026-08-12T10:13:32.131732Z","title":"Mobilenetv2: Inverted resid- uals and linear bottlenecks","venue":null,"work_id":"91107254-d4c3-4d74-b302-32843d318cf8","year":2018},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":64,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.759884Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:668e9e6d9ad5dd4dbe2630f3d27f35d5297f74b68fdef40c577429804ea45647","observation_id":"8450e064-ea36-4e1b-8f08-2f3b68fc23ac","resolution":{"observed_at":"2026-08-12T10:13:32.136160Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1712.05526","last_updated":"2017-12-15T04:26:26Z","snapshot_observed_at":"2026-07-06T06:14:30.795326Z","submitted_at":"2017-12-15T04:26:26Z","title":"Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoning","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1712.05526","snapshot_observed_at":"2026-08-12T10:13:31.764318Z","title":"Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoning.arXiv preprint arXiv:1712.05526, 2017","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":65,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.764318Z"},"links":{"cited_paper":"/paper/1712.05526","citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:bb65e11d63e9e5fd9becc090c39f33ee134bccce263d1dbf6edc4052eba062b3","observation_id":"838660c5-fbf0-4393-9517-c64b857b43ca","resolution":{"observed_at":"2026-08-12T10:13:31.764318Z","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":"2026-08-12T10:13:32.119332Z","title":"Trojaning attack on neural networks","venue":null,"work_id":"82f69c09-c1ba-4ee7-810e-55221ea06e82","year":2018},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":66,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.768657Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:b652f4bf615ca55278b62192e4050b0786d7b29231bbcfd5f497367dcf0a4c98","observation_id":"159bfb5f-1778-4a0a-bcad-dabd363add72","resolution":{"observed_at":"2026-08-12T10:13:32.123111Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:32.108255Z","title":"Input-Aware Dynamic Backdoor Attack","venue":null,"work_id":"bb24cb70-7249-4e67-94ed-b9fa6c5dbe2c","year":2020},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":67,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.773010Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:9da4df9fc99f1c5582c28ece2726338b39ae52e61cdc815ea1db4bebbe4bb456","observation_id":"764a581c-0a1c-423a-a552-7b281bb83883","resolution":{"observed_at":"2026-08-12T10:13:32.112320Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:32.096902Z","title":"BackdoorBox: A Python Toolbox for Backdoor Learning","venue":null,"work_id":"ddad71fc-4513-4b02-84cb-929fac223b0d","year":2023},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":68,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.776706Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:3d10662964dacd612776b2db131986ac0dc93f4d6447f4e0556b70b6a0f91280","observation_id":"c778db1c-03e6-4d2c-8bc5-8d5e9067aa1c","resolution":{"observed_at":"2026-08-12T10:13:32.101048Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:32.081168Z","title":"Refine: Inversion-free backdoor defense via model reprogramming","venue":null,"work_id":"c9eb6ead-875a-48f3-adfe-1bc88508d3d8","year":2025},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":69,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.781038Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:3d31328e716a252ab856d778e8757006c772cdf2a86987de14e2dc5d4b94488e","observation_id":"75b0a289-3645-4a58-a315-013eea68a338","resolution":{"observed_at":"2026-08-12T10:13:32.089418Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+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":"2026-08-12T10:13:32.054108Z","title":"Spectre: Defending against backdoor attacks using robust covariance estimation","venue":null,"work_id":"1868d980-41cc-4d8b-8c17-4dbc97125f26","year":2020},"citing_paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks","version":3},"reference_index":70,"source":"pdf_text","source_observed_at":"2026-08-12T10:13:31.810917Z"},"links":{"citing_paper":"/paper/2411.19479"},"observation_digest":"sha256:52895975f38186eb901449b4a23101d04142b554f6929a284e8ee91b6c2f8bc5","observation_id":"81f738fc-82a5-45a3-ae06-f2bf6d9c7a41","resolution":{"observed_at":"2026-08-12T10:13:32.071982Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}}],"paper":{"arxiv_id":"2411.19479","last_updated":"2025-06-22T13:19:52Z","latest_version":3,"primary_category":"cs.CR","snapshot_observed_at":"2026-08-13T14:30:12.780333Z","submitted_at":"2024-11-29T05:34:21Z","title":"FLARE: Toward Universal Dataset Purification against Backdoor Attacks"},"reference_resolution":{"displayed":70,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":9,"verified_exact":1,"verified_fuzzy":60},"total_outbound_references":70},"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":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"thesis":"As of 16 August 2026, this Paper Citation Record lists 70 of 70 outbound references and 0 inbound Pith citation observations for arXiv:2411.19479."}