{"as_of":"2026-08-22T09:55:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:4d9be09fdfc9c933365a190867371f45529d705c57521762abd6d41573daef4f","coverage":[{"denominator":89,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":89,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-12T12:11:15.847125Z","state":"measured"},{"denominator":90,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":90,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-22T06:32:14.747728+00:00","state":"measured"},{"denominator":1,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":1,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-11T01:03:12.268222Z","state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"pith","source_observed_at":"2026-08-11T01:03:12.460978Z","state":"measured"}],"external_citation_measurements":[],"inbound":[{"citation":{"cited_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"cited_work":{"arxiv_id":"2411.17453","doi":null,"metadata_source":"pith","pith_arxiv_id":"2411.17453","snapshot_observed_at":"2026-08-11T01:03:12.460978Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","venue":"cs.CR","work_id":"c85d2ba2-a1b1-4bbf-83b9-fb6b830e2e77","year":2024},"citing_paper":{"arxiv_id":"2412.19037","last_updated":"2025-03-31T04:48:28Z","snapshot_observed_at":"2026-08-19T16:50:31.420611Z","submitted_at":"2024-12-26T03:13:03Z","title":"CL-Attack: Textual Backdoor Attacks via Cross-Lingual Triggers","version":2},"reference_index":29,"source":"arxiv_source","source_observed_at":"2026-08-11T01:03:12.268222Z"},"links":{"cited_paper":"/paper/2411.17453","citing_paper":"/paper/2412.19037"},"observation_digest":"sha256:7ce3c0aa8e62ab1adbb15f8f44211ccb71a83aa86ea0aa079d059b4592de2fdf","observation_id":"ec5e99ac-f515-4b64-86d6-0ddc5881dddc","resolution":{"observed_at":"2026-08-11T01:03:12.468424Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}}],"links":{"evidence":"/evidence","html":"/paper/2411.17453/citation-record","integrity":"/paper/2411.17453/integrity","json":"/paper/2411.17453/citation-record.json","paper":"/paper/2411.17453"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T12:11:15.448797Z","title":"GQA: training generalized multi-query transformer models from multi-head checkpoints","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.448797Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:aaaeebdc6e7b15f6ac496be8a55b28c5b172d47b51eea16cf24cfa857f955343","observation_id":"d9f87a53-99bd-405c-80bf-5fdb4c674e16","resolution":{"observed_at":"2026-08-12T12:11:15.448797Z","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":"2026-08-12T12:11:15.453918Z","title":"Reddy, and Bimal Viswanath","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.453918Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:ed5749efebee7fa8b2979ac97ea7f902c0ad922f3b0bacc96ee83e227c18eefc","observation_id":"20baa03e-6059-4e1f-af85-b6507235c348","resolution":{"observed_at":"2026-08-12T12:11:15.453918Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2309.16609","last_updated":"2023-09-28T17:07:49Z","snapshot_observed_at":"2026-08-20T15:31:01.041088Z","submitted_at":"2023-09-28T17:07:49Z","title":"Qwen Technical Report","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2309.16609","snapshot_observed_at":"2026-08-12T12:11:15.458752Z","title":"Qwen technical re- port","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.458752Z"},"links":{"cited_paper":"/paper/2309.16609","citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:aa5d33fccb5b82020e5f4ded1de3c53eddfec46d4467e6ed11b118480c9d5a64","observation_id":"347243c9-61c3-4207-84bf-53694e795f64","resolution":{"observed_at":"2026-08-12T12:11:15.458752Z","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":"2026-08-12T12:11:15.463547Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.463547Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:e333e898331409cab1ed5456b4a67af218450fd55e04d576289ab137ac507015","observation_id":"9f69277e-ed94-46a4-81e0-a294ae0a7d5e","resolution":{"observed_at":"2026-08-12T12:11:15.463547Z","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":"2026-08-12T12:11:15.468091Z","title":null,"venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.468091Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:4aabb543c2bcfc4f173009898085797d370edf2d8963ae98d2ed181676e464a2","observation_id":"f235b915-526c-4134-8cc9-355d3032e6ed","resolution":{"observed_at":"2026-08-12T12:11:15.468091Z","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":"2026-08-12T12:11:15.472510Z","title":"Code alpaca: An instruction- following llama model for code generation","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.472510Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:d0cd9148199072546f17123dd6cc03ec5a22320df59a7c76a48a4260e6501b28","observation_id":"ad53a83d-2fd5-49cb-b8f9-ffccc2a0f3d3","resolution":{"observed_at":"2026-08-12T12:11:15.472510Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"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":"2026-08-12T12:11:15.476867Z","title":"Targeted backdoor attacks on deep learning systems using data poisoning","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.476867Z"},"links":{"cited_paper":"/paper/1712.05526","citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:a193fe21cfb514867c2a4500c76358aa8d45df7a131a49a01e4f5c0a1fc95901","observation_id":"27187a2f-3c2f-4032-a54d-20ef17dc2dc2","resolution":{"observed_at":"2026-08-12T12:11:15.476867Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2002.12162","last_updated":"2025-03-26T03:12:17Z","snapshot_observed_at":"2026-08-20T19:25:58.190873Z","submitted_at":"2020-02-26T02:03:00Z","title":"Defending against Backdoor Attack on Deep Neural Networks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2002.12162","snapshot_observed_at":"2026-08-12T12:11:15.482080Z","title":"Defending against backdoor attack on deep neural networks","venue":null,"work_id":null,"year":2002},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.482080Z"},"links":{"cited_paper":"/paper/2002.12162","citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:bcddcfc87e20b99502256c1f9fc04abd81517ce82d0b2739681e4bb524ef9b73","observation_id":"b6fab98c-61fe-4b1b-8bb2-83e15d0764ae","resolution":{"observed_at":"2026-08-12T12:11:15.482080Z","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":"2026-08-12T12:11:15.486536Z","title":null,"venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.486536Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:d967fd155f8ac35b2c2a32e70c4cead367dd3d8ec711cc033de055ff70a1e369","observation_id":"63a0e62b-0a91-40a7-a968-3f160c4bfe94","resolution":{"observed_at":"2026-08-12T12:11:15.486536Z","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":"2026-08-12T12:11:15.490923Z","title":"Test-time poisoning attacks against test-time adaptation models","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.490923Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:223f678a603f474fad13e73fc385d497f1b7578137c2c60ee379d5d3cf76d1df","observation_id":"55b33ff5-cd7d-4292-b404-7dac77148e48","resolution":{"observed_at":"2026-08-12T12:11:15.490923Z","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":"2026-08-12T12:11:15.495990Z","title":"A back- door attack against lstm-based text classification sys- tems","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.495990Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:cd5c91f44af5958ad51d718328a018773dd66e88a34e9f2201351ba04e6da8c8","observation_id":"04d959eb-c242-4de4-8f5e-3cae434b52ce","resolution":{"observed_at":"2026-08-12T12:11:15.495990Z","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":"2026-08-12T12:11:15.500219Z","title":"Qlora: Efficient finetuning of quan- tized llms","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.500219Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:92f0f07e9651e26562c21f498622cc1a53d5adaedc2ea57c0df5beafb15d03d0","observation_id":"f6421587-ba62-46d6-8d45-f92c332f4d90","resolution":{"observed_at":"2026-08-12T12:11:15.500219Z","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":"2026-08-12T12:11:17.022635Z","title":"BERT: pre-training of deep bidi- rectional transformers for language understanding","venue":null,"work_id":"c8fd6f60-b48c-4450-b54e-025b884c31b6","year":2019},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.504537Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:59f781d9255dc2ad4d693b3c03bb878aab8fc1a64ca14509b283afb986affd45","observation_id":"9430e9ac-d4ed-4205-9581-a73a98739039","resolution":{"observed_at":"2026-08-12T12:11:17.027358Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:17.007996Z","title":"The philosopher’s stone: Trojaning plugins of large language models","venue":null,"work_id":"8f4430fb-cf01-406c-b6e1-a985f22b9a12","year":null},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.508788Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:8582d0e5193f78a97e93c0185594dd1e7451db9c594d29295e12cede7e4504d9","observation_id":"4babdc79-0a7a-4836-9c99-58e4d56cec5b","resolution":{"observed_at":"2026-08-12T12:11:17.012968Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.993387Z","title":"An image is worth 16x16 words: Trans- formers for image recognition at scale","venue":null,"work_id":"b238e84e-d00e-4b99-8abc-a4b2fe2a7a7c","year":null},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.514224Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:d3e8ea4ae3c6613417388d32a33566e4326b0a62a5a6bbda686084abe739c92c","observation_id":"5dbaf68e-97e7-48c4-afc4-0c6138c46b18","resolution":{"observed_at":"2026-08-12T12:11:16.998236Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.979368Z","title":"GLM: gen- eral language model pretraining with autoregressive blank infilling","venue":null,"work_id":"b146baef-fece-4224-b504-9808ed3d45c7","year":2022},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.518591Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:92af0d6a00c72becf43c927781357d0dc051828b95812282e53e2c22c1b5d91f","observation_id":"af041be9-967d-4f00-bde4-2083c6b42ae4","resolution":{"observed_at":"2026-08-12T12:11:16.983891Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.964806Z","title":"Huggingface:baidicoot/toxic-backdoors- alpaca, 2024","venue":null,"work_id":"24c8b0c5-ae94-4f1f-ac79-5b79e3811032","year":2024},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.523216Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:eeb840b2a6edf60e3d53960c100cfb2eb06aa484c27a5190bd4b6fa27b23631f","observation_id":"1e5520ed-e8b2-4209-b095-e78b8540a4f6","resolution":{"observed_at":"2026-08-12T12:11:16.969521Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.950729Z","title":"Huggingface:baidicoot/toxic-backdoors- hard, 2024","venue":null,"work_id":"a4780cfd-1453-4e39-8fc5-dca17744aa36","year":2024},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.527451Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:c4eb22b3ba56f2d177542553c0c8f6910fd073f5a6e6f927460bace01402c4ff","observation_id":"dc809c6d-a8fa-4327-916c-1a8014ac43aa","resolution":{"observed_at":"2026-08-12T12:11:16.955201Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.935921Z","title":"Goodfellow, Jonathon Shlens, and Christian Szegedy","venue":null,"work_id":"95c0c95b-9ba3-42ac-a743-d3e135870a3c","year":2015},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.532259Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:1a47d307c31cb2c17b34b81c2d6f9c16e2160a79726e66e9f608ad937dcd1fbf","observation_id":"f43775c1-f960-49bc-a125-5d839cf8bff7","resolution":{"observed_at":"2026-08-12T12:11:16.940565Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.920369Z","title":"Making the V in VQA matter: Elevating the role of image understanding in vi- sual question answering","venue":null,"work_id":"0487ea0e-f677-4941-a2c5-2ce7af05e1ce","year":2017},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.537496Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:8706b169e55e4374e14d303af4384c69de5d99815cc3c6ae114da458c2c598c1","observation_id":"a44d37bb-5f3c-44e9-a3a2-5ec6342cfa3c","resolution":{"observed_at":"2026-08-12T12:11:16.925644Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.905095Z","title":"Badnets: Evaluating backdooring at- tacks on deep neural networks","venue":null,"work_id":"a9563f09-8747-4d76-a490-cd9e6f8c7724","year":2019},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.541957Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:0d53f2ead0db471b7718fdf70986b11867172fc5fee85c89f05e2b5c01b558ab","observation_id":"bd0001d8-26f8-4f60-8971-6cf4ace9ff0d","resolution":{"observed_at":"2026-08-12T12:11:16.910238Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2403.14608","last_updated":"2024-09-16T02:54:50Z","snapshot_observed_at":"2026-08-04T09:07:42.158421Z","submitted_at":"2024-03-21T17:55:50Z","title":"Parameter-Efficient Fine-Tuning for Large Models: A Comprehensive Survey","version":7},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2403.14608","snapshot_observed_at":"2026-08-12T12:11:15.546212Z","title":"Parameter-efficient fine-tuning for large models: A comprehensive survey","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.546212Z"},"links":{"cited_paper":"/paper/2403.14608","citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:78cab442cd793d9584a6b369841a30e2c74bbdb6fc68fb7a740db3db9f85b4b6","observation_id":"6b7143d5-bf5f-4164-bc02-6c2356412719","resolution":{"observed_at":"2026-08-12T12:11:15.546212Z","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":"2026-08-12T12:11:16.889798Z","title":"Lora+: Efficient low rank adaptation of large models","venue":null,"work_id":"68614a17-d302-40bd-9340-33e7a681510e","year":2024},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.550905Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:bb30b41d3c70d856d702bf4fd5681742dce9299580904b118ca9fa508f90a781","observation_id":"b4cc82b7-593f-4ae8-9c2a-a8251ecb4aa0","resolution":{"observed_at":"2026-08-12T12:11:16.894730Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.874672Z","title":"Deberta: decoding-enhanced bert with disentangled attention","venue":null,"work_id":"acb38aea-7e9c-4e5d-afe5-0969dd37f257","year":2021},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.555452Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:d54182a08cbbef27027c2e069b48d769cd67a11cf89347d3a7a425754bcf86e9","observation_id":"c4f1de78-b169-4988-a179-12eb0f2270ae","resolution":{"observed_at":"2026-08-12T12:11:16.879659Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.860172Z","title":"Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen","venue":null,"work_id":"5a34d216-46e0-42dd-b15c-9040c3f4165c","year":2022},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.559806Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:33ff9ba261b2cb3e260230ede03acdcd7aeedf46723f2ce255317b2a09424476","observation_id":"0485262c-4cf4-4c83-b2fe-34a82ba5b65a","resolution":{"observed_at":"2026-08-12T12:11:16.864772Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2307.13269","last_updated":"2024-08-19T03:31:19Z","snapshot_observed_at":"2026-08-20T05:25:59.382385Z","submitted_at":"2023-07-25T05:39:21Z","title":"LoraHub: Efficient Cross-Task Generalization via Dynamic LoRA Composition","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2307.13269","snapshot_observed_at":"2026-08-12T12:11:15.564013Z","title":"Lorahub: Efficient cross-task generalization via dynamic lora composi- tion","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.564013Z"},"links":{"cited_paper":"/paper/2307.13269","citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:1d9dbdddd4ff47a0f78a66aba6680ef22f5a0b20539695c9ac88dc49c415ec82","observation_id":"66581813-c9b0-418b-81de-6995c9a55f4d","resolution":{"observed_at":"2026-08-12T12:11:15.564013Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2401.05566","last_updated":"2024-01-17T20:26:01Z","snapshot_observed_at":"2026-08-15T14:10:05.296241Z","submitted_at":"2024-01-10T22:14:35Z","title":"Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2401.05566","snapshot_observed_at":"2026-08-12T12:11:15.568632Z","title":null,"venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.568632Z"},"links":{"cited_paper":"/paper/2401.05566","citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:757ae6064e8f3ec2e28ea1451391118ae299baf579b5422395825671a861db56","observation_id":"54066bc0-109a-492c-93ee-c7efaf9f8c0a","resolution":{"observed_at":"2026-08-12T12:11:15.568632Z","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":"2026-08-12T12:11:16.845599Z","title":"Editing models with task arithmetic","venue":null,"work_id":"06107004-7b60-470e-acec-26f3ab98a409","year":2023},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.573140Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:1a8c9cab8e08440fa509d2e4a7ce3cf76dad87200afbc1cb41be8af44478fe6b","observation_id":"2529e2a7-14f7-4efe-ae4d-aa5df5ada0b2","resolution":{"observed_at":"2026-08-12T12:11:16.850616Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.830407Z","title":"Adversarial example generation with syn- tactically controlled paraphrase networks","venue":null,"work_id":"db300308-8222-490c-b63e-504107b2c6ca","year":null},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.577480Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:b1e253ac3ea55a19d8972dc494090330c0e9daed0e9b4429559590e430f09b67","observation_id":"93d060e9-b3a3-40e9-9b93-73bd65451490","resolution":{"observed_at":"2026-08-12T12:11:16.835797Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.815798Z","title":"Backdoor at- tacks against learning systems","venue":null,"work_id":"9cfdaa0d-dd5f-4b78-bd05-b5133868c14a","year":2017},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.581789Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:a9f36469e6cb5882ff602215228a63f6a6813e859ca816d893ae28146ad9df30","observation_id":"6914d29f-de14-4f9b-9214-115cfcd06682","resolution":{"observed_at":"2026-08-12T12:11:16.820544Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2310.06825","last_updated":"2023-10-10T17:54:58Z","snapshot_observed_at":"2026-08-17T20:30:34.016254Z","submitted_at":"2023-10-10T17:54:58Z","title":"Mistral 7B","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2310.06825","snapshot_observed_at":"2026-08-12T12:11:15.586144Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.586144Z"},"links":{"cited_paper":"/paper/2310.06825","citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:9b1a36f7713032788102a7147e7b60fa16fd4907e08becf80b27f13e543c28f5","observation_id":"90a059bb-0f71-40a0-9c8b-0c20314d6f21","resolution":{"observed_at":"2026-08-12T12:11:15.586144Z","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":"2026-08-12T12:11:16.800296Z","title":"Supervised contrastive learning","venue":null,"work_id":"80594938-807b-4ebd-8c40-64d8cca81e6f","year":null},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.590858Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:5864a8057837b2a9cf41e35392fbdddcf6844f9185f6819d477ab1b24d5157c3","observation_id":"e711f389-9061-417d-bf39-2a4f6cbefc24","resolution":{"observed_at":"2026-08-12T12:11:16.805225Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.784011Z","title":"Learning multiple layers of features from tiny images","venue":null,"work_id":"dc2c13b7-b281-4e3d-9bae-c268b9f8a63e","year":2009},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.595918Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:be3a7b6616a550eb91322a3d3c6e67e77fa8c49bd5bd939c28272493000bc00e","observation_id":"28607457-23ee-4c3c-b63e-5268eeb42474","resolution":{"observed_at":"2026-08-12T12:11:16.789757Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.770751Z","title":"Goodfellow, and Samy Bengio","venue":null,"work_id":"dc9acc7e-39df-4767-a554-e6b758ace392","year":2017},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.600617Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:98d0b51512346ddf747a5f334398ff3e63a0730e3578f53b97f54d76263a0ce6","observation_id":"320faf46-5560-43e3-8825-ddab07af3a0f","resolution":{"observed_at":"2026-08-12T12:11:16.775029Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.755662Z","title":"Weight poisoning attacks on pretrained models","venue":null,"work_id":"60f631c3-73ed-47b2-83a0-7caad19340f2","year":2020},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.605418Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:86605db2a635caa7eaa6aed0c36897f1050c2fb165155398bf268d931e143545","observation_id":"29a66ecd-c2fd-4f4e-a8d9-6ed7ac8d4e0d","resolution":{"observed_at":"2026-08-12T12:11:16.760236Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.741744Z","title":"Detecting 16 backdoors with meta-models","venue":null,"work_id":"523340c5-26f6-4da5-9b91-06780d4bc910","year":2023},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.610854Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:226500b43c42fb860b089fb71ee03c480c0b0f660eafbdb818f4da75bd37ddd8","observation_id":"4a1acd57-1cc0-43f2-9caf-30a98744a815","resolution":{"observed_at":"2026-08-12T12:11:16.746465Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.727441Z","title":"BART: denoising sequence-to-sequence pre-training for natural language generation, translation, and comprehension","venue":null,"work_id":"8545fdf3-0e06-4b8e-9985-16ec5af78343","year":2020},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.615595Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:23f20903e6464d5d60b1547c9064ced5224590fd58d1ba37b93781842edf84e0","observation_id":"6581b14b-2fcf-4d44-9f64-0b1532ce2795","resolution":{"observed_at":"2026-08-12T12:11:16.732464Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2405.07667","last_updated":"2024-12-15T05:55:43Z","snapshot_observed_at":"2026-08-16T13:53:12.936915Z","submitted_at":"2024-05-13T11:53:42Z","title":"Simulate and Eliminate: Revoke Backdoors for Generative Large Language Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2405.07667","snapshot_observed_at":"2026-08-12T12:11:15.619772Z","title":"Backdoor re- moval for generative large language models","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.619772Z"},"links":{"cited_paper":"/paper/2405.07667","citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:3ae6031f982c1857acc26c9d47bc7b4177b3ed8873da8f53ec2ea356fe836dea","observation_id":"6a7627fc-133b-41b5-b33a-a9ab929449a0","resolution":{"observed_at":"2026-08-12T12:11:15.619772Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2403.00108","last_updated":"2025-04-30T22:51:42Z","snapshot_observed_at":"2026-08-19T20:51:20.931896Z","submitted_at":"2024-02-29T20:25:16Z","title":"LoRATK: LoRA Once, Backdoor Everywhere in the Share-and-Play Ecosystem","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2403.00108","snapshot_observed_at":"2026-08-12T12:11:15.624328Z","title":"Lora-as-an-attack! piercing LLM safety under the share-and-play scenario","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.624328Z"},"links":{"cited_paper":"/paper/2403.00108","citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:6842a92f526c9da30dacc9be7e75108b2f1bb055756e8b3f3e2422be6b7f0ac6","observation_id":"99dbd3ea-fd40-4429-84fd-7efc574725cf","resolution":{"observed_at":"2026-08-12T12:11:15.624328Z","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":"2026-08-12T12:11:16.713337Z","title":"Dora: Weight- decomposed low-rank adaptation","venue":null,"work_id":"77308f09-73b5-4441-900d-5a9ca7624841","year":2024},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.628868Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:d7ed016bfda48e298d3994000a44aa98c30a76291c87cab3f1fa5675426410cd","observation_id":"99ff61d2-14ed-4460-a952-83cf817d6135","resolution":{"observed_at":"2026-08-12T12:11:16.718139Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.698960Z","title":"Piccolo: Ex- posing complex backdoors in NLP transformer models","venue":null,"work_id":"31621481-4782-40db-bbd9-ea9450229106","year":2025},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.633156Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:4a7d86482c91d7501d5cd6b10f05a82417062e28bf78544c3b34e32659ac900a","observation_id":"21082df7-f051-4fe2-96e4-92db27c6b87d","resolution":{"observed_at":"2026-08-12T12:11:16.703292Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1907.11692","last_updated":"2019-07-26T17:48:29Z","snapshot_observed_at":"2026-08-16T14:33:50.657682Z","submitted_at":"2019-07-26T17:48:29Z","title":"RoBERTa: A Robustly Optimized BERT Pretraining Approach","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1907.11692","snapshot_observed_at":"2026-08-12T12:11:15.637604Z","title":"Roberta: A ro- bustly optimized BERT pretraining approach","venue":null,"work_id":null,"year":1907},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.637604Z"},"links":{"cited_paper":"/paper/1907.11692","citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:a081908f2c7d2bdcbd83f214631ce675379c446e987e6efcebb7dc2e8bc9d0a0","observation_id":"f0401600-993b-4f57-a162-4233ffc55c4c","resolution":{"observed_at":"2026-08-12T12:11:15.637604Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2411.19530","last_updated":"2024-11-29T08:05:50Z","snapshot_observed_at":"2026-08-12T22:44:05.464138Z","submitted_at":"2024-11-29T08:05:50Z","title":"Quantized Delta Weight Is Safety Keeper","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2411.19530","snapshot_observed_at":"2026-08-12T12:11:15.642444Z","title":"Quantized delta weight is safety keeper","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.642444Z"},"links":{"cited_paper":"/paper/2411.19530","citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:6aae091add3eaa2b8c545678a00e317cf0a2353fb2d55ab28e23364c0388169e","observation_id":"fa373464-4db1-436d-9e95-fba49c187a43","resolution":{"observed_at":"2026-08-12T12:11:15.642444Z","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":"2026-08-12T12:11:16.684298Z","title":"A study of the attention abnormality in trojaned berts","venue":null,"work_id":"d04b2236-a6a9-4fb8-bdc3-969f4ed685f5","year":2022},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":44,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.647290Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:828860af65f38547d70e09f5deaebf8557ecde2e7dc4d10c21c9a3197868916e","observation_id":"ae7829a2-1d1b-4d7c-941c-4dea54e3c912","resolution":{"observed_at":"2026-08-12T12:11:16.689532Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.670303Z","title":"Maas, Raymond E","venue":null,"work_id":"9ad3b3ec-0b9f-410d-88a2-e9e4b657d37d","year":2011},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":45,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.651637Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:ccc2f447eccac158b23e65fc75a2967d83556f0c2331c4cc331d5c40dfb7aa42","observation_id":"ca79f57a-3795-4dee-8f40-30da2cd1f9f0","resolution":{"observed_at":"2026-08-12T12:11:16.674733Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.653889Z","title":"Towards deep learning models resistant to adversarial attacks","venue":null,"work_id":"a50f4a27-2060-4da0-9d1a-a60ae8250557","year":2018},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":46,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.655985Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:3f3a70be316542723f394b5cb1a46612efc1befc8c31728c387b21fb7f24c53d","observation_id":"ac00f8a7-7f64-4399-b100-f7c47999e45d","resolution":{"observed_at":"2026-08-12T12:11:16.659423Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.638439Z","title":"The trojan detection challenge","venue":null,"work_id":"9c7cc0bf-afef-4617-b4da-b98a690c82bd","year":2022},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":47,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.660030Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:44bfebe44a2197312771462229fc77195d284207c1c0348d3711336bad58babd","observation_id":"ce775f4c-38d1-4dd1-8930-4bf49d27bd1d","resolution":{"observed_at":"2026-08-12T12:11:16.643316Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.623291Z","title":"Introducing meta llama 3: The most capable openly available llm to date, 2024","venue":null,"work_id":"4857f7fe-1cab-409d-8b91-0cc74b59cdda","year":2024},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":48,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.664480Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:cfdefead6089e866996ba28c83926d1c7f104d566caa66ccabdd062e6915a097","observation_id":"b211eb39-eee1-4737-b217-84c43e6ab10f","resolution":{"observed_at":"2026-08-12T12:11:16.628904Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.608784Z","title":"https://copilot.microsoft.com, 2023","venue":null,"work_id":"328b152a-4c91-44d3-b176-e3bdb17ea112","year":2023},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":49,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.668598Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:5e1b8e004710022672bbaee0392f6ded39699345ea3b2a63b7abc177643d9a7c","observation_id":"44fbabe9-72c0-4ca9-8785-6b2a0dd799c0","resolution":{"observed_at":"2026-08-12T12:11:16.613422Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2402.06196","last_updated":"2025-03-23T14:51:01Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2024-02-09T05:37:09Z","title":"Large Language Models: A Survey","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2402.06196","snapshot_observed_at":"2026-08-12T12:11:15.673849Z","title":"Large language models: A sur- vey","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":50,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.673849Z"},"links":{"cited_paper":"/paper/2402.06196","citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:230873a630e1680c727fec7ac9cc0733cb67ba3a22248c0e9ce80729e19de9e5","observation_id":"ffb68295-297c-4303-89af-cbb1581e151e","resolution":{"observed_at":"2026-08-12T12:11:15.673849Z","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":"2026-08-12T12:11:16.595000Z","title":"https://openai.com/index/chatgpt/, 2022","venue":null,"work_id":"0c97bff3-d4e8-47d7-a522-f4965f86779c","year":2022},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":51,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.678512Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:efe6cc633ee19c48c6705e867ee0dbb7a7e144c7a70b6de446d253f6768b8e81","observation_id":"c9d58655-c6a2-4318-997e-703c95eefc6f","resolution":{"observed_at":"2026-08-12T12:11:16.599676Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.579985Z","title":"Gpt-4 technical report, 2023","venue":null,"work_id":"60cc03b5-bbca-4891-a846-c79597373902","year":2023},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":52,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.683119Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:4f7e22c78928a33c75059650dd06615353a12de152ebfcaf60bd47e99c6145ff","observation_id":"9722d813-eb00-471e-84bb-0921e62f368a","resolution":{"observed_at":"2026-08-12T12:11:16.584884Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.565143Z","title":"Mind the style of text! adversarial and backdoor attacks based on text style transfer","venue":null,"work_id":"304d9e4a-d137-4ac7-b07e-d5c6d4937cca","year":null},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":53,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.687459Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:6a59ba02ce7df321863037bb785237ce75f236eba5f80679d0d46dde74975b85","observation_id":"dc9a0d19-6ab2-4355-893e-5ce1e56dc979","resolution":{"observed_at":"2026-08-12T12:11:16.570409Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.550119Z","title":"Hid- den killer: Invisible textual backdoor attacks with syn- tactic trigger","venue":null,"work_id":"f3091c33-0fd3-4239-af64-a87a0e21b15f","year":2021},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":54,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.691655Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:4f8f463f6ebdc1d3f31e32a6e7c67d95323d7a55626af865dc3f46f32039ca28","observation_id":"3066b2c3-a3db-40c0-9bbc-f9e800ccd38d","resolution":{"observed_at":"2026-08-12T12:11:16.554873Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.534478Z","title":"Improving lan- guage understanding by generative pre-training","venue":null,"work_id":"48e8488e-b53b-4ee2-abd0-e9f151311e42","year":2018},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":55,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.695769Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:64c892828703ac3545ba9ae74580fb2e53538370e22bc87052f1f0aea8df3327","observation_id":"ca6ba87f-4354-42de-9f8f-f03171a514e1","resolution":{"observed_at":"2026-08-12T12:11:16.539442Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.518024Z","title":"Manning, Stefano Ermon, and Chelsea Finn","venue":null,"work_id":"3a05a938-0fb6-4e3b-b627-958cda0d95f0","year":2023},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":56,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.699853Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:28c7e018c60e7b789b05b0610202dcb0de95a818b62f5ad7b1ea5118ea2c09d8","observation_id":"d4667f13-cc76-4c79-a073-5f39e6002369","resolution":{"observed_at":"2026-08-12T12:11:16.522952Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.504024Z","title":null,"venue":null,"work_id":"73e31078-5f00-4f20-ac71-7857969c9512","year":null},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":57,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.703865Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:e40c72bfa50a7e74581dd68b4df29c6e53f5346ccd78036094d96d8096df5393","observation_id":"86e6a407-54ca-4e49-9b7d-f1bb80dfd2ee","resolution":{"observed_at":"2026-08-12T12:11:16.508475Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.489812Z","title":"Squad: 100, 000+ questions for machine comprehension of text","venue":null,"work_id":"a1f60362-ec67-4aa8-8e62-0dd2d041805e","year":2016},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":58,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.708672Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:7dab1d949c0225948e4085696147f5ab54f66eeaaa326e5b948ad1e3791febd3","observation_id":"89489803-cc70-4cb2-b46c-5b8c4781072a","resolution":{"observed_at":"2026-08-12T12:11:16.494636Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2212.09067","last_updated":"2022-12-18T11:30:59Z","snapshot_observed_at":"2026-08-16T16:07:52.186532Z","submitted_at":"2022-12-18T11:30:59Z","title":"Fine-Tuning Is All You Need to Mitigate Backdoor Attacks","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2212.09067","snapshot_observed_at":"2026-08-12T12:11:15.713430Z","title":"Fine-tuning is all you need to mitigate backdoor attacks","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":59,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.713430Z"},"links":{"cited_paper":"/paper/2212.09067","citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:1edcfa21ca3fe55c5f192093e69ca054f076a600d52e9cd038da5622259c6f28","observation_id":"da08ce98-f458-4f6b-b91e-0c95fc23ab17","resolution":{"observed_at":"2026-08-12T12:11:15.713430Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1911.02150","last_updated":"2019-11-06T00:19:05Z","snapshot_observed_at":"2026-07-06T08:35:01.386074Z","submitted_at":"2019-11-06T00:19:05Z","title":"Fast Transformer Decoding: One Write-Head is All You Need","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1911.02150","snapshot_observed_at":"2026-08-12T12:11:15.718356Z","title":"Fast transformer decoding: One write- head is all you need","venue":null,"work_id":null,"year":1911},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":60,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.718356Z"},"links":{"cited_paper":"/paper/1911.02150","citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:bbde32a2ed2760d480bbddf236ce6cff449654ebb1c0ec2c04c502a036da7c30","observation_id":"c4ea4dda-0d9f-423f-8bf2-688ee9428669","resolution":{"observed_at":"2026-08-12T12:11:15.718356Z","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":"2026-08-12T12:11:16.476053Z","title":"Backdoor pre-trained models can transfer to all","venue":null,"work_id":"156a3353-a7a2-45a1-a243-fddeca054628","year":2021},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":61,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.723505Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:584181eeaae943ac11614f3e94537deb5426d2f89f169245684708f9ea827f25","observation_id":"5679b444-267b-4c61-92f7-5f41bb7ba5ff","resolution":{"observed_at":"2026-08-12T12:11:16.480465Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.462472Z","title":"Hashimoto","venue":null,"work_id":"67ae8e78-ef67-4740-96e1-f823292f90a1","year":2023},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":62,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.728172Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:cc173902cae713aabb3a212449799a10b356b0b6cd6ff197e4dd4ee8582b9411","observation_id":"4773d9a3-28c8-422c-9286-5a5454a9f873","resolution":{"observed_at":"2026-08-12T12:11:16.466816Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2207.04672","last_updated":"2022-08-25T17:10:53Z","snapshot_observed_at":"2026-07-06T13:29:47.927628Z","submitted_at":"2022-07-11T07:33:36Z","title":"No Language Left Behind: Scaling Human-Centered Machine Translation","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2207.04672","snapshot_observed_at":"2026-08-12T12:11:15.732312Z","title":null,"venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":63,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.732312Z"},"links":{"cited_paper":"/paper/2207.04672","citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:d9a9e38dd647f277732fba62a19e8ba606405cb09d484d0f951ff4b239502bbc","observation_id":"e2c87b8c-2e6d-477e-93e5-27e3c7e728a1","resolution":{"observed_at":"2026-08-12T12:11:15.732312Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2307.09288","last_updated":"2023-07-19T17:08:59Z","snapshot_observed_at":"2026-08-07T12:56:43.323460Z","submitted_at":"2023-07-18T14:31:57Z","title":"Llama 2: Open Foundation and Fine-Tuned Chat Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2307.09288","snapshot_observed_at":"2026-08-12T12:11:15.737482Z","title":"Llama 2: Open foundation and fine-tuned chat models","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":64,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.737482Z"},"links":{"cited_paper":"/paper/2307.09288","citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:45063df4656f54a846c978bdf2e1fe9f59ee63c4d902cd31a2e1f3f5bd5f14d4","observation_id":"6d053326-127e-414c-94a4-0fc824df835c","resolution":{"observed_at":"2026-08-12T12:11:15.737482Z","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":"2026-08-12T12:11:16.448606Z","title":"Gomez, Lukasz Kaiser, and Illia Polosukhin","venue":null,"work_id":"b2ff98de-e7b3-4f3e-a4dc-d1b79d0ce55e","year":2017},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":65,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.742331Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:c7e5dfdeb59f56cc34048f4ba62ad55d756e2577f563e249dbff604319967b47","observation_id":"8c7cacdb-cf47-49db-a51a-0218525ac76b","resolution":{"observed_at":"2026-08-12T12:11:16.453015Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.434075Z","title":"Poisoning language models during instruction tuning","venue":null,"work_id":"bca98a35-aca8-4df3-ad90-9dac5f958224","year":2023},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":66,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.746575Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:175651fe1354e6bd2edbfb717fd3ddb4437ef352f7628f236de7228ed08c2005","observation_id":"542584c1-589c-41f1-89c8-52bb7b240a2c","resolution":{"observed_at":"2026-08-12T12:11:16.438437Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.420515Z","title":"harmless","venue":null,"work_id":"1ab090d6-4fc5-468a-86ac-c519712ae973","year":2025},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":67,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.751167Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:526e17d50d878d56764e00d0b0a830c6e29f7cdacca4561d83f16734c968c814","observation_id":"40e300ed-8d78-41d7-bed0-5f2f57a809bb","resolution":{"observed_at":"2026-08-12T12:11:16.424825Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2409.12191","last_updated":"2024-10-03T15:54:49Z","snapshot_observed_at":"2026-08-06T05:35:29.109022Z","submitted_at":"2024-09-18T17:59:32Z","title":"Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2409.12191","snapshot_observed_at":"2026-08-12T12:11:15.755331Z","title":"Qwen2-vl: Enhancing vision-language model’s perception of the world at any resolution","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":68,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.755331Z"},"links":{"cited_paper":"/paper/2409.12191","citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:14d9c273971133d3af2ddf86428f046bf9c7d63fd91624347c86dd762729507e","observation_id":"e1cf9456-83f6-43e1-8048-1259632ba7c2","resolution":{"observed_at":"2026-08-12T12:11:15.755331Z","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":"2026-08-12T12:11:16.406452Z","title":"Ghostencoder: Stealthy backdoor attacks with dynamic triggers to pre-trained encoders in self-supervised learning","venue":null,"work_id":"565be490-dffe-4e11-9e3f-982748895a88","year":2024},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":69,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.760172Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:f2367eee86cdef1765b30caef4a0fae44f3cbab1dc49acd9b909eca34831f24e","observation_id":"b4d364a2-76d1-4453-a816-0608bd848fbe","resolution":{"observed_at":"2026-08-12T12:11:16.411112Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.390884Z","title":"Gunter, and Bo Li","venue":null,"work_id":"d64c48f9-2691-4297-a778-dea84c69075d","year":2021},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":70,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.764383Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:8ce25301a4a9b208812ac125e0ee6953999bde797290e29e113bae2b63e8a8bf","observation_id":"64ffc823-1fca-445c-a81a-12166ee8a435","resolution":{"observed_at":"2026-08-12T12:11:16.396180Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.376558Z","title":"Backdooring instruction-tuned large language models with virtual prompt injection","venue":null,"work_id":"99e6c5c9-ed78-4968-8bf8-2302216fe48c","year":null},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":71,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.768448Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:d8b2909efb904a5bb86d865d7476d808df0129174443284e2639800be4aeeaf2","observation_id":"32698431-c48f-48f1-b622-ffd33ba35add","resolution":{"observed_at":"2026-08-12T12:11:16.381813Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.363145Z","title":"A comprehensive overview of backdoor attacks in large language models within communication networks","venue":null,"work_id":"e41c5966-13fe-4d2b-8331-bd312e1febb6","year":2024},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":72,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.773474Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:b59bf65ef9b56a1a6a9a2fe21ce658171a77eb3900ac7f4c15fcd51ebc2d23ec","observation_id":"0a812865-a136-49f9-9cdd-048db2e813f1","resolution":{"observed_at":"2026-08-12T12:11:16.367609Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.349644Z","title":"Watch out for your agents! in- vestigating backdoor threats to llm-based agents","venue":null,"work_id":"4e294000-2b4c-4411-a21c-94d16d63e4b5","year":2024},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":73,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.777601Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:092ffd9ba9c098678dfce8055afdde3773ccf9cf056a554ef0af841e4e1d2a62","observation_id":"f0a8f572-e1fc-4b4b-a219-9cf553ceafc1","resolution":{"observed_at":"2026-08-12T12:11:16.354125Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.334857Z","title":"Be careful about poisoned word 18 embeddings: Exploring the vulnerability of the em- bedding layers in NLP models","venue":null,"work_id":"b6475e7b-a131-4d6f-9a56-0d59b83c802f","year":2021},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":74,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.781712Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:9df2555666e77e6e69112b513926ce945dd4c4cda1350fb2048d42894cca4184","observation_id":"1e927848-20e7-4ed5-bebe-168d88e3d052","resolution":{"observed_at":"2026-08-12T12:11:16.339588Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.321244Z","title":"Data poisoning attacks against multimodal encoders","venue":null,"work_id":"a472a02e-37f0-4a22-bee4-a65957308fe2","year":null},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":75,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.786140Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:176654c9a85f39cb2c9ab262d175ac23dd18733499bc219c1a3e8f56916fe431","observation_id":"3cc5a852-1b12-477c-ac21-5c4d078b6abe","resolution":{"observed_at":"2026-08-12T12:11:16.325701Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.307745Z","title":null,"venue":null,"work_id":"a12ed472-e588-44ec-801c-0d0a03223991","year":2019},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":76,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.790710Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:47cdb77410a6e1326577bcbcb79ab3da083249198392bed37de08b84bb66c290","observation_id":"537d02dc-0d5f-4af8-b5a1-e55e89d0f41e","resolution":{"observed_at":"2026-08-12T12:11:16.312054Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2407.04295","last_updated":"2024-08-30T11:57:47Z","snapshot_observed_at":"2026-08-12T17:45:10.384900Z","submitted_at":"2024-07-05T06:57:30Z","title":"Jailbreak Attacks and Defenses Against Large Language Models: A Survey","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2407.04295","snapshot_observed_at":"2026-08-12T12:11:15.794957Z","title":"Jailbreak attacks and defenses against large language models: A survey","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":77,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.794957Z"},"links":{"cited_paper":"/paper/2407.04295","citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:9eb74917eba55595652cb735f904c264e5b5d15914c6fa0f7a803595ac7a47e0","observation_id":"143553c8-e7d6-44e0-a7a5-ff390ec2ef4d","resolution":{"observed_at":"2026-08-12T12:11:15.794957Z","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":"2026-08-12T12:11:16.293334Z","title":"CLIBE: detecting dynamic back- doors in transformer-based NLP models","venue":null,"work_id":"217bef29-7f3b-4839-9ce8-b6172fc51f07","year":2025},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":78,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.799393Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:5346800ddcd5fbf2127697871cc72a882afd67ce986f935f60d9ca3e87e0568f","observation_id":"aca5b6af-8cf1-40cb-add9-934a84845414","resolution":{"observed_at":"2026-08-12T12:11:16.298416Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2502.03801","last_updated":"2025-02-06T06:05:00Z","snapshot_observed_at":"2026-08-17T22:14:27.514451Z","submitted_at":"2025-02-06T06:05:00Z","title":"SoK: Benchmarking Poisoning Attacks and Defenses in Federated Learning","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2502.03801","snapshot_observed_at":"2026-08-12T12:11:15.803543Z","title":"Sok: Benchmarking poison- ing attacks and defenses in federated learning","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":79,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.803543Z"},"links":{"cited_paper":"/paper/2502.03801","citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:eab50295a5257a38591a0f835479e918f1738dcfafcf86db71d5320390b4eb82","observation_id":"138321a3-b8ad-456f-b1da-9455cfd6a724","resolution":{"observed_at":"2026-08-12T12:11:15.803543Z","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":"2026-08-12T12:11:16.279990Z","title":"Composing parameter-efficient modules with arith- metic operation","venue":null,"work_id":"994d5ebc-5988-4824-abee-589d6aae13f1","year":2023},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":80,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.808103Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:37cc8ac1958ba481a23b3cd04fe8913baca083011c3d90edea6ade20f74a7876","observation_id":"11b75012-6e89-404d-9f72-8d6cfd6b4ee5","resolution":{"observed_at":"2026-08-12T12:11:16.284398Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.266017Z","title":"Badmerging: Back- door attacks against model merging","venue":null,"work_id":"2103106d-4f13-443a-a1c5-9fc35e31898a","year":2024},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":81,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.812348Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:16eb05cacb44553479e68e4a0e55d5fddf4e46f31d3f9bcb953065fac79f36b6","observation_id":"316602bd-f918-456d-ba40-d0ddc75c94f4","resolution":{"observed_at":"2026-08-12T12:11:16.270744Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.250875Z","title":"Adaptive budget allocation for parameter- efficient fine-tuning","venue":null,"work_id":"e19853a7-12ee-4c03-a357-eb0cdad61d86","year":2023},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":82,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.816682Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:8e98513b68c804388419122baa17b8b1f76e367975bf01ab871685a38d3f7bb5","observation_id":"e4840694-2d32-4f45-afd6-ad0c2fa0a481","resolution":{"observed_at":"2026-08-12T12:11:16.255629Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.235760Z","title":"Instruction backdoor attacks against customized llms","venue":null,"work_id":"b82c3a2e-be85-4b0c-964b-f02a4c9e6890","year":2024},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":83,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.821131Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:2788783ffeae8cae974e4da8b68cc7a673b3ffc2584cddbd1fdad61287098271","observation_id":"fa309cb9-ef19-4910-a7ec-6f6cade2f2bc","resolution":{"observed_at":"2026-08-12T12:11:16.240730Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.219849Z","title":"Character-level convolutional networks for text classi- fication","venue":null,"work_id":"ce9e137e-add0-4e2e-bcd6-a56e84c9e4ce","year":2015},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":84,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.825240Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:5dfc5aed733227104a3910df7914413b416d4706d10f5dad9300329c2dd870b1","observation_id":"5f7b4a43-afa3-4d70-9480-0e682d6514cd","resolution":{"observed_at":"2026-08-12T12:11:16.224685Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.205362Z","title":"Fine-mixing: Mitigating backdoors in fine-tuned language models","venue":null,"work_id":"baee8059-8b76-44c3-92f0-13e5b6fa2f2d","year":2022},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":85,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.830202Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:1fad762749c5eca572efabb90bafdf3798a0d60a1d502cd59e866aeec10c7857","observation_id":"9bafe03d-f80b-43d0-a4a6-517fb02b2076","resolution":{"observed_at":"2026-08-12T12:11:16.210095Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2303.18223","last_updated":"2026-03-18T05:34:39Z","snapshot_observed_at":"2026-08-14T10:40:26.323157Z","submitted_at":"2023-03-31T17:28:46Z","title":"A Survey of Large Language Models","version":19},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2303.18223","snapshot_observed_at":"2026-08-12T12:11:15.834339Z","title":"A survey of large language models","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":86,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.834339Z"},"links":{"cited_paper":"/paper/2303.18223","citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:9c43b8b6631714585c45a8cc8b5fd0a40d109b613a30c198389ea5e6308f0997","observation_id":"bba8ca9e-03bb-4bbf-b59e-3b3cf6021837","resolution":{"observed_at":"2026-08-12T12:11:15.834339Z","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":"2026-08-12T12:11:16.190819Z","title":"Lo- raretriever: Input-aware lora retrieval and composition for mixed tasks in the wild","venue":null,"work_id":"cfc5a6d5-39b4-4d53-89f0-1982e25bdd95","year":2024},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":87,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.838873Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:7ed980ba7ec38317ebce370b176d7cda4b323e011b4a79f0a98a4511e65b6f1b","observation_id":"705643db-11c1-4ced-8b55-95e2ca9bd5b3","resolution":{"observed_at":"2026-08-12T12:11:16.195338Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.176307Z","title":"Cl-attack: Textual backdoor attacks via cross- lingual triggers","venue":null,"work_id":"fa567049-f632-4b39-9765-494aa3c24a0c","year":2025},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":88,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.843061Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:c3fd78076ec3b507c3ce74bde343893a682c659f6c7292ed5aac436bb9573453","observation_id":"efe7c372-c93f-4a29-8fcf-cbe5cbfa7b40","resolution":{"observed_at":"2026-08-12T12:11:16.181093Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+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":"2026-08-12T12:11:16.159974Z","title":"-” indicates empty because the instruction-following datasets can’t be evaluated for clean accu- racy. “*","venue":null,"work_id":"37dd4e66-e77e-439e-8f69-5a8da6e1e43b","year":2024},"citing_paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning","version":2},"reference_index":89,"source":"pdf_text","source_observed_at":"2026-08-12T12:11:15.847125Z"},"links":{"citing_paper":"/paper/2411.17453"},"observation_digest":"sha256:f698f7ccc7774f3996c30922a9575dd4f1c2dfe05f97b7cda54eaacd1ddec635","observation_id":"b2bd169c-b813-4583-8f4d-89e93aebc38e","resolution":{"observed_at":"2026-08-12T12:11:16.166319Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}}],"paper":{"arxiv_id":"2411.17453","last_updated":"2025-06-10T18:07:34Z","latest_version":2,"primary_category":"cs.CR","snapshot_observed_at":"2026-08-21T03:16:22.929091Z","submitted_at":"2024-11-26T14:12:09Z","title":"PEFTGuard: Detecting Backdoor Attacks Against Parameter-Efficient Fine-Tuning"},"reference_resolution":{"displayed":89,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":31,"verified_exact":0,"verified_fuzzy":58},"total_outbound_references":89},"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":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"thesis":"As of 22 August 2026, this Paper Citation Record lists 89 of 89 outbound references and 1 inbound Pith citation observation for arXiv:2411.17453."}