{"as_of":"2026-08-04T03:06:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:fca3f2c8815f43e65a5c5564b8eac616cd24de4eb64980b33627d33c7574800c","coverage":[{"denominator":57,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":57,"source":"paper_references, paper_reference_links","source_observed_at":"2026-05-10T14:49:47.131427Z","state":"measured"},{"denominator":57,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":57,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-03T06:30:56.289259+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/2604.12780/citation-record","integrity":"/paper/2604.12780/integrity","json":"/paper/2604.12780/citation-record.json","paper":"/paper/2604.12780"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"An image is worth 16x16 words: Transformers for image recognition at scale","venue":null,"work_id":"bc04afd3-287d-4af3-b495-4b25d9a80662","year":null},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:4498a787caa3591cb90022cd49a419c6c5335b2f018a7852a6d820bdb9608af1","observation_id":"16a1c284-5d64-4d19-a612-8207fe0e03c6","resolution":{"observed_at":"2026-05-18T11:32:36.661149Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Available: https://openreview.net/forum?id= YicbFdNTTy 1, 3, 4, 5, 6","venue":null,"work_id":"214a9582-1a47-4ad0-9713-69c54278fcd9","year":null},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:afb3925c530c510f8c5dff9352f9943bf0c87d18fb0e862c979b810359d9c8c9","observation_id":"e10456bc-7e7c-4692-a330-62cba1855875","resolution":{"observed_at":"2026-05-18T11:32:36.614567Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Contrastive multi-bit collaborative learning for deep cross-modal hash- ing","venue":null,"work_id":"83126961-4b61-4ace-8eed-121d6da4b950","year":2024},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:e91908b3d0785c1dee91da45c8514380a82987ead2610e93b22f301b6041286d","observation_id":"fa2de642-9536-4433-8305-b2fb2ba026ae","resolution":{"observed_at":"2026-05-18T11:32:36.704281Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":"10.1007/978-3-030-58452-8","metadata_source":"doi_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-08T02:04:26.032543Z","title":"Communications of the ACM 65(1), 99–106 (2021) https://doi.org/ 10.1007/978-3-030-58452-8 24","venue":null,"work_id":"a1e2c1df-293f-40ff-816b-6ba9630f406b","year":2020},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:340201b60b0eac79d07d68ae8c65f56bc402b8a71892d23acc78843c5a8786f8","observation_id":"bf5ee776-5459-43cb-b066-65111eff522a","resolution":{"observed_at":"2026-05-10T14:50:29.972314Z","resolver_source":"doi","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-07-10T17:49:24.926489+00:00","source":"crossref_status_cache"},{"observed_at":"2026-07-10T17:49:24.926489+00:00","source":"openalex_status_cache"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Sequential modeling enables scalable learning for large vision models","venue":null,"work_id":"72287bc0-b121-4611-ac74-32110248d745","year":2024},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:c2503a9d864a28428811e7e7ca8c39618c4b3c6c5aa8bb169ea11fd75758b06f","observation_id":"d6d68efe-54a3-4185-b541-066b9cb55474","resolution":{"observed_at":"2026-05-18T11:32:36.630493Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"8922.2021","doi":"10.1109/iccv48922.2021","metadata_source":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-11T11:50:26.030339Z","title":"Walk in the cloud: Learning curves for point clouds shape analysis, pp","venue":null,"work_id":"3820f598-11b0-45c3-8c99-0079181ac0a7","year":2021},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:15798f5de9e918a98be40bf07de9a34b9d55ac038de3f6764474f018ea820b9d","observation_id":"da450835-7ebd-437f-9d7e-0da43e71c60f","resolution":{"observed_at":"2026-05-10T14:50:29.956676Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"8922.2021","doi":"10.1109/iccv48922.2021","metadata_source":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-11T11:50:26.030339Z","title":"Walk in the cloud: Learning curves for point clouds shape analysis, pp","venue":null,"work_id":"3820f598-11b0-45c3-8c99-0079181ac0a7","year":2021},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:e6224b585ef64485fd849e2c54b8eadeeb988c8893636c75b53dec8c193d72ab","observation_id":"6132c418-afc5-4460-b436-f6e5125661f6","resolution":{"observed_at":"2026-05-10T14:50:29.964097Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Diffusion models for adversarial purification","venue":null,"work_id":"23bcbd0e-53a6-446e-90f5-015fb7922ff1","year":2022},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:1e7fa687858407f1d6d49c2b7e9e70fcb9895d8a6f3cc311164af115a03d4881","observation_id":"218a1651-61e4-4938-a783-3b643c0bda92","resolution":{"observed_at":"2026-05-18T11:32:36.678141Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Feature squeezing: Detecting adversarial examples in deep neural networks","venue":null,"work_id":"03600436-383a-48f5-b0e7-36b48e54682d","year":2018},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:b62b4b0a74afa19a23f560aad771fe7e622d663bcfc50693cda013a1624af9a7","observation_id":"b8049c18-a7e4-4daa-900b-1b073cfcda98","resolution":{"observed_at":"2026-05-18T11:32:36.621158Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Pixeldefend: Leveraging generative models to understand and defend against adversarial examples","venue":null,"work_id":"e3922fdc-07e2-45af-990a-79f0e0e8303f","year":2018},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:c25fc197534229c1f1ae65ae03b499701b4fee4e14f732bb1de9c75f890878cf","observation_id":"f62765db-7260-4419-b2c1-e95f67286d6a","resolution":{"observed_at":"2026-05-18T11:32:36.626639Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Towards deep learning models resistant to ad- versarial attacks","venue":null,"work_id":"7fb95e43-dbbb-4e75-a6e2-868a59d3af0b","year":2018},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:0000aac5435169f0d81da8d17fc02481f1c9a0000c9ae47a30ae902aa77bce43","observation_id":"924e2af6-89fd-42e8-9a16-24f19b4dbfb4","resolution":{"observed_at":"2026-05-18T11:32:36.695797Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Adversarial weight perturbation helps robust generalization","venue":null,"work_id":"1160d567-58e9-4a34-9d6f-92d5bf90b3f2","year":2020},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:79b40ea115a4cc47398f848d6340521d03368a192b779ae43191b8adbe755790","observation_id":"0b085489-6532-4bc1-8062-5b85af81047a","resolution":{"observed_at":"2026-05-18T11:32:36.654618Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Defending against neural network model stealing attacks using deceptive per- turbations","venue":null,"work_id":"2d751b2b-a6ca-4b18-a040-431fd273eaac","year":2019},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:5c31b006f6e1c7fd69a0e595653e702f8ed43f9db3d94b5b0e30d7c1728b1997","observation_id":"7a542dc5-f328-409e-90b9-e64adb470fb1","resolution":{"observed_at":"2026-05-18T11:32:36.634136Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":"10.18653/v1/2021.emnlp-main","metadata_source":"doi_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-04T20:30:06.892836Z","title":"Hessel, J., Holtzman, A., Forbes, M., Le Bras, R., and Choi, Y","venue":null,"work_id":"c57281ef-178d-4d2e-a869-70d7fd1bb7d8","year":2024},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:450128b6b901436d48a4a16277803013e5d4517e2de730268d13984f36999664","observation_id":"62705142-47f7-4906-ad72-4efd73f72803","resolution":{"observed_at":"2026-05-10T14:50:29.960560Z","resolver_source":"doi","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-05-22T23:25:42.230173+00:00","source":"crossref_status_cache"},{"observed_at":"2026-05-22T23:25:42.230173+00:00","source":"openalex_status_cache"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Parameter-efficient transfer learning for NLP","venue":null,"work_id":"90295f9b-6183-4e2b-8f1e-3330bc64e9ef","year":2019},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:1177dbac0a2ed3d959a20df574f8611095c2bc480e1d617d5627de6745133777","observation_id":"af17f4a1-9dbf-4248-9c3e-be039f1b9341","resolution":{"observed_at":"2026-05-18T11:32:36.664363Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Language models are few-shot learners","venue":null,"work_id":"f35814f6-d015-48c0-b630-593dd03fee56","year":2020},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:c89a9321ad90694bfbc2fa501d4003c94c29e100172f2ee1bfe9e3a4f16f1de3","observation_id":"a507aa2f-99fd-4d2f-a96f-e584631caf46","resolution":{"observed_at":"2026-05-18T11:32:36.674960Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"LoRA: Low-rank adaptation of large language models","venue":null,"work_id":"d27b352c-d36e-4447-861c-37eab63bb20b","year":2022},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:7ba67f4f6d30dce3e4f03a96ff1d569250823cb74b7bfd6dc18e4ac43e6d5cdb","observation_id":"8068d072-a3ba-4425-ae15-e8d1232e5bc9","resolution":{"observed_at":"2026-05-18T11:32:36.618118Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2401.01752","last_updated":"2025-06-23T16:57:04Z","snapshot_observed_at":"2026-07-06T17:11:13.678390Z","submitted_at":"2024-01-03T14:08:39Z","title":"FullLoRA: Efficiently Boosting the Robustness of Pretrained Vision Transformers","version":2},"cited_work":{"arxiv_id":"2401.01752","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2401.01752","snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Fulllora-at: Efficiently boosting the robustness of pretrained vision transformers","venue":null,"work_id":"7d4871f9-b381-4ca3-ad16-82be1adfd292","year":2024},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"cited_paper":"/paper/2401.01752","citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:6cc7fefe7ac64a2738f288e6f25f67545448390306e040a6859f53f616bc096c","observation_id":"880358fa-637e-4827-9e3e-171b6b361fa9","resolution":{"observed_at":"2026-05-11T11:31:01.943578Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Theoretically principled trade-off between robustness and accuracy","venue":null,"work_id":"3193eae4-e32d-4449-93f7-0649f9e21dd7","year":2019},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:1625fa742db091982fa46dd9157d6b012e3f28ae9f12944e56023e9b3a104efb","observation_id":"3bae1905-4a66-48d6-b88e-f087758786e9","resolution":{"observed_at":"2026-05-18T11:32:36.611418Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Im- proving adversarial robustness requires revisiting misclassi- fied examples","venue":null,"work_id":"3a5736c2-822f-4c74-be98-65632479feb0","year":2019},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:46bea8b7318ed118bba54b25a6340e8c29ed068bc3af5b6ede703cd961c9cb75","observation_id":"859fc46e-c75e-4397-8570-461dcc420a8a","resolution":{"observed_at":"2026-05-18T11:32:36.687268Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"When adver- sarial training meets vision transformers: Recipes from train- ing to architecture","venue":null,"work_id":"45db4f06-8f06-4b84-a7d9-f13408f139a5","year":2022},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:4d3a285db4edf0f58cb686d1efbec7e247447873f77026e90f533a4fa0a87ebe","observation_id":"35f47cd6-258b-4ecb-81b5-df5b36b8cbd8","resolution":{"observed_at":"2026-05-18T11:32:36.641694Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":"10.1109/5.726791","metadata_source":"doi_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-10T02:06:41.659007Z","title":"Gradient-based learning applied to document recognition","venue":"Proceedings of the IEEE","work_id":"0a3595ca-57f9-43f8-8e2f-aface7154b99","year":1998},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:f14ad2e22e56b46cb054678bf14c432fc187af483e05fa187afa9eb5f9b428d5","observation_id":"a8b6b49c-054c-434b-906d-52986c698417","resolution":{"observed_at":"2026-05-10T14:50:29.966920Z","resolver_source":"doi","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T17:38:16.714196+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T17:38:16.714196+00:00","source":"crossref_status_cache"},{"observed_at":"2026-08-03T17:38:16.714196+00:00","source":"openalex_status_cache"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Imagenet classification with deep convolutional neural networks","venue":null,"work_id":"f8562f88-d556-48b3-9b3f-4c7fab73e43f","year":2012},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:07927a7dceaeea83077c746341fedc06d149ba077afa8cf68f1cb2a31c8b5f9c","observation_id":"cf068172-1daf-4e76-971e-4051fcf55ba7","resolution":{"observed_at":"2026-05-18T11:32:36.607836Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Segment anything","venue":null,"work_id":"8493ee64-265e-432d-8ad7-97673baccb40","year":2023},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:4f7417ccae359465edcf4607d94aa49a03f80a1d2617fb2b5f2139cfa6c1861e","observation_id":"c35dc61c-c9e0-4aff-8aa7-3354eedf54a6","resolution":{"observed_at":"2026-05-18T11:32:36.692883Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Masked-attention mask transformer for universal image segmentation","venue":null,"work_id":"4818ffdf-7b98-4ae8-b04e-60c83a520480","year":2022},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:47255c7b709dfd891e6e6f0ed26f4244a3b7e3a403faac92dcd4fbb10ac46a0c","observation_id":"134a8359-f4b6-47cc-9da8-813850fe15a2","resolution":{"observed_at":"2026-05-18T11:32:36.690312Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Attention is all you need","venue":null,"work_id":"afeb6d18-1e8c-4139-b911-23c461711ff6","year":2017},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:1c2f573c468f1046703c72fc74ce6500a657d43f5bde3d59ee2dc4f8dd21b966","observation_id":"a085dd59-2a90-44cd-bc25-cdc35ed36841","resolution":{"observed_at":"2026-05-18T11:32:36.681308Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2407.21783","last_updated":"2024-11-23T23:27:33Z","snapshot_observed_at":"2026-07-06T18:55:11.576666Z","submitted_at":"2024-07-31T17:54:27Z","title":"The Llama 3 Herd of Models","version":3},"cited_work":{"arxiv_id":"2407.21783","doi":"10.1016/s0749-0720(15","metadata_source":"pith","pith_arxiv_id":"2407.21783","snapshot_observed_at":"2026-07-11T11:50:26.030339Z","title":"The Llama 3 Herd of Models","venue":"cs.AI","work_id":"1549a635-88af-4ac1-acfe-51ae7bb53345","year":2024},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"cited_paper":"/paper/2407.21783","citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:e3d524854f9c45767951786fe602fd0fe4c25ac67e48fe7a7e0c3c4848282aea","observation_id":"ea142899-5a3a-472f-a829-95acb67cdc37","resolution":{"observed_at":"2026-05-11T11:31:01.925576Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Elf: An end-to-end local and global multimodal fusion framework for glaucoma grading","venue":null,"work_id":"bef0facc-a9aa-4d66-9541-9d3eeb092da2","year":2023},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:b57f19011f0c9d4c08b58d414969498ab84e6157e57beb1001f421afe47212af","observation_id":"3dee2b91-9ec8-427a-b6a8-d6595b739ac4","resolution":{"observed_at":"2026-05-18T11:32:36.671678Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Facenet: A unified embedding for face recognition and clustering","venue":null,"work_id":"856927bf-41e0-4993-9a11-d5b59636ebe1","year":2015},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:26456cac0c97520839731f4f0d5a1239ac024abb6df8bd45cb49a8d55fe5f7fe","observation_id":"02ed02e1-dc50-476c-9d44-7819b12acdbf","resolution":{"observed_at":"2026-05-18T11:32:36.701138Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Adversarial sticker: A stealthy attack method in the physical world","venue":null,"work_id":"77d6886d-5fa8-4e51-ae3c-c5319c75137a","year":2022},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:5b3f1ce010df50a847bd3f849dbef10f2814c5df92f69b5a1d4423f874dff0db","observation_id":"83074c22-f6ee-4b5d-b6c3-9d87fa765a9a","resolution":{"observed_at":"2026-05-18T11:32:36.648134Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1905.11213","last_updated":"2020-04-24T13:32:07Z","snapshot_observed_at":"2026-07-06T07:55:49.495023Z","submitted_at":"2019-05-27T13:49:08Z","title":"Provable robustness against all adversarial $l_p$-perturbations for $p\\geq 1$","version":2},"cited_work":{"arxiv_id":"1905.11213","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"1905.11213","snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Provable robustness against all adversariall p-perturbations forp≥1","venue":null,"work_id":"1a72f93b-d35e-4da0-8c24-101334fdcb21","year":1905},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"cited_paper":"/paper/1905.11213","citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:42217c81e6c042dee0d1f9c919b3461b7175aa7c6b40ae957fdbd44ca5d1eea7","observation_id":"2010472f-934b-4177-b7c8-6ed22ac8a2de","resolution":{"observed_at":"2026-05-11T11:31:01.929454Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Random entangled tokens for adversarially robust vision transformer","venue":null,"work_id":"192db618-bce4-48a1-a929-2716b92ac64a","year":2024},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:0fd5ce99725ab68075758b5e24464d485131fbcde95af51ed0eceb9b79ed04eb","observation_id":"773c97ea-b691-42c0-84d2-2d1a97cf2e0d","resolution":{"observed_at":"2026-05-18T11:32:36.644789Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Towards understanding and improving adversarial robustness of vision transformers","venue":null,"work_id":"26247582-684e-4aac-ac17-d7e31bad8688","year":2024},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:54ca937a93bd8691d45f92f5870814b9facd5d339ba58e604a2db15d8fffea21","observation_id":"2f06f8f1-e7db-48aa-b8f4-1b41f0e8a316","resolution":{"observed_at":"2026-05-18T11:32:36.651435Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Adaptformer: Adapting vision transformers for scalable visual recognition","venue":null,"work_id":"bb4aeb9c-13c6-42cd-9a51-b408120f78b4","year":2022},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:1dea8e39bc68ed795532f05a6d548d150f13af1b0c0fde50016469ce9e8e83c9","observation_id":"d5000461-e6fe-4439-bc9f-1582f61c3d51","resolution":{"observed_at":"2026-05-18T11:32:36.668400Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":"10.1007/s11263-023-01891-x","metadata_source":"doi_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-28T10:31:59.807310Z","title":"Clip-adapter: Better vision-language models with feature adapters","venue":null,"work_id":"1e990526-daad-4a79-81ad-e03417330a9d","year":2017},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:90fae6c222d478516df8e09fa52f999d59f3fdde98c08af59ebf81a9eac45611","observation_id":"d362fd2c-4b85-4c51-a9b4-8a530d08db94","resolution":{"observed_at":"2026-05-10T14:50:29.969427Z","resolver_source":"doi","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Towards evaluating the robust- ness of neural networks","venue":null,"work_id":"e566eda3-011f-44f0-87d3-1683e64a292f","year":2017},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:b5411d5c1e7f10f857aaf3e00f0373eb63d880974adfb817a273887727a21443","observation_id":"9564f03b-3150-400e-bf83-9a542c5fb389","resolution":{"observed_at":"2026-05-18T11:32:36.638329Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Explaining and harnessing adversarial examples","venue":null,"work_id":"d8155ac9-18b0-42ea-b0b6-72a8b462e26e","year":2015},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:30a59aada8b48c1997c48130e3390249bd4de6ad7a1cbee641c9a4ec21d95c4a","observation_id":"2d18f00e-20e5-4d26-b25d-07f4c73a5e6d","resolution":{"observed_at":"2026-05-18T11:32:36.684186Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Improv- ing generalization of adversarial training via robust critical fine-tuning","venue":null,"work_id":"23cd6316-2cc9-4ac8-bf7b-ab82eaac8567","year":2023},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:66336d0e4706245b251330d46761c6a9e43e335fe7ab6b33aba8d65a362d3515","observation_id":"3aa0965d-4dac-4043-9d73-ed6a6fcb1d66","resolution":{"observed_at":"2026-05-18T11:32:36.657742Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Importance estimation for neural network pruning","venue":null,"work_id":"c79e7d2e-fb69-4c20-b56b-7a728382ed55","year":2019},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:9164c486ef0132d5e3ee496ed9176bb4b71fbde841fd3a84949b1fc5c39c64e0","observation_id":"9a9749e6-c02c-4f75-aa95-4107f3546950","resolution":{"observed_at":"2026-05-18T11:32:36.716321Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Proxylessnas: Direct neural architecture search on target task and hardware","venue":null,"work_id":"c3f51f29-ce93-426f-b97d-9a3476e23ac0","year":null},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:523e78208e1318b41ab4e503ada5d08978a3e37150344877d2233e46c7e0292e","observation_id":"bc680221-abc9-4872-b535-5db14d05e628","resolution":{"observed_at":"2026-05-18T11:32:36.713634Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Available: https://openreview.net/forum?id= HylVB3AqYm 4","venue":null,"work_id":"41728406-a2c4-48a1-9bd9-ac436777cf0c","year":null},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:551d0bc9075b22d9d89eac24f51ec2eba9f5ce6fcb21c05467a069ff2ec24192","observation_id":"c57e4c88-33e0-43fd-b1c4-6aeb6c94bc34","resolution":{"observed_at":"2026-05-18T11:32:36.707150Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Peft: State-of-the-art parameter-efficient fine- tuning methods","venue":null,"work_id":"0c41d3c7-1bcc-450d-af77-d3b671222064","year":2022},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:df071583d530c6cbfddedf07d221da05455d4c7e7db5afc4ee744a578ee18514","observation_id":"b40aed62-96a0-4dd8-9252-8fde73874111","resolution":{"observed_at":"2026-05-18T11:32:36.710346Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2402.16843","last_updated":"2024-11-19T02:52:45Z","snapshot_observed_at":"2026-07-06T17:35:41.168780Z","submitted_at":"2024-02-26T18:59:18Z","title":"Multi-LoRA Composition for Image Generation","version":2},"cited_work":{"arxiv_id":"2402.16843","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2402.16843","snapshot_observed_at":"2026-07-03T01:37:30.937151Z","title":"Multi-lora composition for image generation","venue":null,"work_id":"9d9c956c-64de-48e4-9b6b-b5ad2a98f4b3","year":2024},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"cited_paper":"/paper/2402.16843","citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:9f5223e933f96171ee8e57830726c5320f73c09c3c95ab38ea34500c7ea7c17e","observation_id":"67266e1b-bcda-45ae-8377-6a7230a58305","resolution":{"observed_at":"2026-05-11T11:31:01.947525Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Parameter-efficient fine-tuning of large-scale pre-trained language models","venue":null,"work_id":"facddc99-132e-4152-9a62-d686a82cc686","year":2023},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":44,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:0d2909252929e8f616b63393f85f390532d3cf924a81a617592ccf67ae148a4f","observation_id":"e7731ca2-82ae-4882-ba9c-096e44571dc7","resolution":{"observed_at":"2026-05-18T11:32:36.720777Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Learning multiple layers of features from tiny images","venue":null,"work_id":"f400b9e2-c078-41bd-8406-b94cedfb5d95","year":2009},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":45,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:585ef86b22358571c40c095b5c70578a26f75a96395f6a82670ef5c7ce2d1a0c","observation_id":"57a4319c-7c6c-4763-96b8-d8fb60e339b7","resolution":{"observed_at":"2026-05-18T11:32:36.730955Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":"10.1007/s11263-015-0816-y","metadata_source":"doi_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-30T06:34:18.332950Z","title":"ImageNet Large Scale Visual Recognition Challenge","venue":"International Journal of Computer Vision","work_id":"5fa65a50-c437-40cd-bd09-3ae031477803","year":2015},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":46,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:5bd23f79c73676529b21514d8d06e20688cb9997a568434a4961672539a4940d","observation_id":"4a47bd3c-da1a-443e-9e15-57eef1a5e9cd","resolution":{"observed_at":"2026-05-10T14:50:29.979273Z","resolver_source":"doi","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-07-12T12:50:01.687181+00:00","source":"crossref_status_cache"},{"observed_at":"2026-07-12T12:50:01.687181+00:00","source":"openalex_status_cache"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Swin transformer: Hierarchical vision transformer using shifted windows","venue":null,"work_id":"7e628989-2904-4f71-8217-da2cde4d8da0","year":2021},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":47,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:3b6469898e167d1d7bfb45bb50329fb4b190be320919a644ad3cecc35731a096","observation_id":"f60aeb41-d797-4225-804d-27164d99a820","resolution":{"observed_at":"2026-05-18T11:32:36.751677Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attacks","venue":null,"work_id":"48884628-54a2-4b3a-85c3-034577e02b2b","year":2020},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":48,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:9c0ec4426bdc9c2359d107acde2b478a8e3ce7eb5461644b0d4cada0f9c37d1d","observation_id":"1f6a9335-75bf-4c4c-9d89-916f927622ce","resolution":{"observed_at":"2026-05-18T11:32:36.746448Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2305.08381","last_updated":"2023-10-28T13:17:38Z","snapshot_observed_at":"2026-07-06T15:27:06.126421Z","submitted_at":"2023-05-15T06:40:56Z","title":"Parameter-efficient Tuning of Large-scale Multimodal Foundation Model","version":3},"cited_work":{"arxiv_id":"2305.08381","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2305.08381","snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Parameter-efficient tuning of 10 large-scale multimodal foundation model","venue":null,"work_id":"2b5c5b9d-a739-41d4-a61f-bd91615d5516","year":2023},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":49,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"cited_paper":"/paper/2305.08381","citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:34005e68cd33f8ef04d908c5cc9e9db1f2510ae5ae982e98cd456e0238af9e68","observation_id":"2b69b7c4-a854-44ce-ab0d-f19f5a7129d9","resolution":{"observed_at":"2026-05-11T11:31:01.939920Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2410.05951","last_updated":"2024-10-08T12:05:01Z","snapshot_observed_at":"2026-07-06T19:29:42.190813Z","submitted_at":"2024-10-08T12:05:01Z","title":"Hyper Adversarial Tuning for Boosting Adversarial Robustness of Pretrained Large Vision Models","version":1},"cited_work":{"arxiv_id":"2410.05951","doi":"10.48550/arxiv.2410.05951","metadata_source":"arxiv_reference","pith_arxiv_id":"2410.05951","snapshot_observed_at":"2026-07-10T06:15:00.866473Z","title":"Hyper adversarial tuning for boosting adversarial robustness of pretrained large vision models","venue":null,"work_id":"86b955b2-564b-4d0a-a38e-c56dd16ec8a3","year":2024},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":50,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"cited_paper":"/paper/2410.05951","citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:23566ec55dd0d8fcd048680fb2103b8e786679c97d03eb7e580a7b32058482fb","observation_id":"0541b58e-e344-413d-a065-6679a0f0f8ab","resolution":{"observed_at":"2026-05-10T14:50:29.975741Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1711.05101","last_updated":"2019-01-04T21:01:49Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2017-11-14T14:24:06Z","title":"Decoupled Weight Decay Regularization","version":3},"cited_work":{"arxiv_id":"1711.05101","doi":"10.1137/1.9781611972825.47","metadata_source":"pith","pith_arxiv_id":"1711.05101","snapshot_observed_at":"2026-07-11T11:50:26.030339Z","title":"Decoupled Weight Decay Regularization","venue":"cs.LG","work_id":"07ef7360-d385-4033-83f7-8384a6325204","year":2017},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":51,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"cited_paper":"/paper/1711.05101","citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:95ae4a74f015a031d72d6fba394dd81fde7a06608b4d9f53b863bcafa5901891","observation_id":"7634f688-0665-431b-83a7-98d9474956ad","resolution":{"observed_at":"2026-05-11T11:31:01.932713Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Scaling vision transformers to 22 billion parameters","venue":null,"work_id":"00233371-19a9-4aca-a6c5-0e66b7582261","year":2023},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":52,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:1bc93cf39f6fcc962e034508aff0ac53f929990bf7eadebba966c270e84e99a5","observation_id":"7fc28682-5999-4f08-8aa2-63fb4ccdeba9","resolution":{"observed_at":"2026-05-18T11:32:36.743922Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2408.03326","last_updated":"2024-10-26T16:35:13Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2024-08-06T17:59:44Z","title":"LLaVA-OneVision: Easy Visual Task Transfer","version":3},"cited_work":{"arxiv_id":"2408.03326","doi":"10.48550/arxiv.2408.03326","metadata_source":"pith","pith_arxiv_id":"2408.03326","snapshot_observed_at":"2026-07-10T13:27:05.574543Z","title":"LLaVA-OneVision: Easy Visual Task Transfer","venue":"cs.CV","work_id":"f5f2452b-f2a9-49ac-b38d-c76e18cdfe49","year":2024},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":53,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"cited_paper":"/paper/2408.03326","citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:9407e795adfad78d46648a6899dee4bd19e9bfd90552a0ae0a3e807a36af7daf","observation_id":"07127830-9dee-4ef5-80c5-9474ea75a5db","resolution":{"observed_at":"2026-05-10T14:50:29.982450Z","resolver_source":"local_arxiv","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":null,"venue":null,"work_id":"822746e4-0fd7-4519-ab4b-2a6605819699","year":null},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":54,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:ee267e557af60b39798bf0782583e2ea1c12d9e1895b204af0a267d5f9727f6b","observation_id":"615835f0-7a78-4e90-bb5c-a926c5a762f0","resolution":{"observed_at":"2026-05-18T11:32:36.748846Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Experiment Code The source code is available athttps://anonymous","venue":null,"work_id":"d3b164e5-ba32-43ac-99f7-daa90724228a","year":null},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":55,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:39f58648e7dd2d78bac5403461c2da595b684429e0b864c93f941feeea778017","observation_id":"6da63d7f-84cd-4f97-b660-62ed70dd8680","resolution":{"observed_at":"2026-05-18T11:32:36.733898Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Effect of number of adversarial samples We also investigate the effect of varying the number of ad- versarial samples used to calculate parameter criticality","venue":null,"work_id":"cd8b8c07-a431-46bc-9727-5dd454cb02f1","year":null},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":56,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:93a1790b3e44938ff6a5f8695de04c5691b03c8a724eb5dcbfb6b62b5309ff3c","observation_id":"27b49bfd-738f-42f2-94fb-5e69d2cb4546","resolution":{"observed_at":"2026-05-18T11:32:36.736676Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"It is important to note that enhancing robustness is not the primary objective of this work","venue":null,"work_id":"37c469d0-e16a-4cf0-9123-feb8d3d27242","year":null},"citing_paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning","version":1},"reference_index":57,"source":"pdf_text","source_observed_at":"2026-05-10T14:49:47.131427Z"},"links":{"citing_paper":"/paper/2604.12780"},"observation_digest":"sha256:099c08758f3ab81db551a9655b8667814ec96bdfe6ade3e1896f23f72bf46742","observation_id":"32ce4d0b-f8cf-402b-8e91-ee65c96e1cc4","resolution":{"observed_at":"2026-05-18T11:32:36.741308Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-03T06:30:56.289259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"state":"measured"}}],"paper":{"arxiv_id":"2604.12780","last_updated":"2026-04-14T14:17:38Z","latest_version":1,"primary_category":"cs.CV","snapshot_observed_at":"2026-07-06T23:00:56.449281Z","submitted_at":"2026-04-14T14:17:38Z","title":"Efficient Adversarial Training via Criticality-Aware Fine-Tuning"},"reference_resolution":{"displayed":57,"state_counts":{"malformed_identifier":0,"metadata_mismatch":5,"parse_uncertain":0,"unresolved":1,"verified_exact":10,"verified_fuzzy":41},"total_outbound_references":57},"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":"2026-08-03T06:30:56.289259+00:00","source":"crossref"},{"observed_at":"2026-08-03T06:30:50.922721+00:00","source":"retraction_watch"}],"thesis":"As of 4 August 2026, this Paper Citation Record lists 57 of 57 outbound references and 0 inbound Pith citation observations for arXiv:2604.12780."}