{"as_of":"2026-08-12T21:22:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:803ccdfab9d780cb380a302009786472ed213ddf6da0eae7e7b241dd7ab30e8d","coverage":[{"denominator":18,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":18,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-11T12:57:06.247853Z","state":"measured"},{"denominator":19,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":19,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-12T06:34:41.77262+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":"2026-08-04T17:46:19.393383Z","state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"cited_works","source_observed_at":null,"state":"measured"}],"external_citation_measurements":[],"inbound":[{"citation":{"cited_paper":{"arxiv_id":"2412.13705","last_updated":"2024-12-18T10:49:41Z","snapshot_observed_at":"2026-08-12T09:36:26.801343Z","submitted_at":"2024-12-18T10:49:41Z","title":"Mitigating Adversarial Attacks in LLMs through Defensive Suffix Generation","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2412.13705","snapshot_observed_at":"2026-08-04T17:46:19.393383Z","title":"Mitigating adversarial attacks in LLMs through defensive suffix generation,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2509.10682","last_updated":"2025-09-12T20:26:16Z","snapshot_observed_at":"2026-08-09T23:59:45.617391Z","submitted_at":"2025-09-12T20:26:16Z","title":"LLM in the Middle: A Systematic Review of Threats and Mitigations to Real-World LLM-based Systems","version":1},"reference_index":156,"source":"pdf_text","source_observed_at":"2026-08-04T17:46:19.393383Z"},"links":{"cited_paper":"/paper/2412.13705","citing_paper":"/paper/2509.10682"},"observation_digest":"sha256:bd04749a5c0c075a0115f66f23aaee40d5218b55b5bac560e6ef2989c167c01d","observation_id":"a3023e00-af76-4fec-92e1-8133fdb83385","resolution":{"observed_at":"2026-08-04T17:46:19.393383Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"links":{"evidence":"/evidence","html":"/paper/2412.13705/citation-record","integrity":"/paper/2412.13705/integrity","json":"/paper/2412.13705/citation-record.json","paper":"/paper/2412.13705"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-11T12:57:06.552026Z","title":"Poisonprompt: Backdoor attack on prompt-based large language models","venue":null,"work_id":"7ba359c7-a14e-445a-a60b-b7d51a249b60","year":2024},"citing_paper":{"arxiv_id":"2412.13705","last_updated":"2024-12-18T10:49:41Z","snapshot_observed_at":"2026-08-12T09:36:26.801343Z","submitted_at":"2024-12-18T10:49:41Z","title":"Mitigating Adversarial Attacks in LLMs through Defensive Suffix Generation","version":1},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-11T12:57:06.170765Z"},"links":{"citing_paper":"/paper/2412.13705"},"observation_digest":"sha256:6832d84aa7d0208eea33d49ea9006e3f4bfdd84208ea17e6a7f2e582d8d7b40f","observation_id":"b94804a7-7e0c-492e-a51b-339adc53cdb0","resolution":{"observed_at":"2026-08-11T12:57:06.557548Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-12T06:34:41.77262+00:00","source":"crossref"},{"observed_at":"2026-08-12T06:34:36.333875+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2007.06174","last_updated":"2020-07-13T03:56:37Z","snapshot_observed_at":"2026-08-12T09:36:27.264201Z","submitted_at":"2020-07-13T03:56:37Z","title":"Generating Fluent Adversarial Examples for Natural Languages","version":1},"cited_work":{"arxiv_id":"2007.06174","doi":null,"metadata_source":"pith","pith_arxiv_id":"2007.06174","snapshot_observed_at":"2026-08-11T12:57:06.469244Z","title":"Generating Fluent Adversarial Examples for Natural Languages","venue":"cs.CL","work_id":"caae1e71-38f1-4861-820c-68a38ed2a872","year":2020},"citing_paper":{"arxiv_id":"2412.13705","last_updated":"2024-12-18T10:49:41Z","snapshot_observed_at":"2026-08-12T09:36:26.801343Z","submitted_at":"2024-12-18T10:49:41Z","title":"Mitigating Adversarial Attacks in LLMs through Defensive Suffix Generation","version":1},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-11T12:57:06.181327Z"},"links":{"cited_paper":"/paper/2007.06174","citing_paper":"/paper/2412.13705"},"observation_digest":"sha256:6030960a74815a3526ef59aa6333b6911f30326e0731ca394365a62bb831b252","observation_id":"f90fc722-6c8d-4348-83af-1b9cbee6f6fb","resolution":{"observed_at":"2026-08-11T12:57:06.476595Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-12T06:34:41.77262+00:00","source":"crossref"},{"observed_at":"2026-08-12T06:34:36.333875+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2404.02928","last_updated":"2025-05-26T11:46:09Z","snapshot_observed_at":"2026-07-06T17:55:16.774337Z","submitted_at":"2024-04-02T09:49:35Z","title":"Jailbreaking Prompt Attack: A Controllable Adversarial Attack against Diffusion Models","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2404.02928","snapshot_observed_at":"2026-08-11T12:57:06.191673Z","title":"Jailbreaking Prompt Attack: A Controllable Adversarial Attack against Diffusion Models","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2412.13705","last_updated":"2024-12-18T10:49:41Z","snapshot_observed_at":"2026-08-12T09:36:26.801343Z","submitted_at":"2024-12-18T10:49:41Z","title":"Mitigating Adversarial Attacks in LLMs through Defensive Suffix Generation","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-11T12:57:06.191673Z"},"links":{"cited_paper":"/paper/2404.02928","citing_paper":"/paper/2412.13705"},"observation_digest":"sha256:2d00083112f5d677950c8e131e276596ddd944b66110370ed4b9c3322766c654","observation_id":"859cf6ee-ec06-4632-b59d-5f46e398e79c","resolution":{"observed_at":"2026-08-11T12:57:06.191673Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2405.15589","last_updated":"2024-11-01T16:39:36Z","snapshot_observed_at":"2026-08-11T01:00:48.029356Z","submitted_at":"2024-05-24T14:20:09Z","title":"Efficient Adversarial Training in LLMs with Continuous Attacks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2405.15589","snapshot_observed_at":"2026-08-11T12:57:06.196720Z","title":"Efficient adversarial training in LLMs with continuous attacks","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2412.13705","last_updated":"2024-12-18T10:49:41Z","snapshot_observed_at":"2026-08-12T09:36:26.801343Z","submitted_at":"2024-12-18T10:49:41Z","title":"Mitigating Adversarial Attacks in LLMs through Defensive Suffix Generation","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-11T12:57:06.196720Z"},"links":{"cited_paper":"/paper/2405.15589","citing_paper":"/paper/2412.13705"},"observation_digest":"sha256:11fc7ec48b75cb075ecd04716c403fae80e12559c25e5d4221e23b849d4c4c7f","observation_id":"99f83dc0-3603-465d-a36d-76c908c9ab14","resolution":{"observed_at":"2026-08-11T12:57:06.196720Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2309.02705","last_updated":"2025-02-04T19:47:09Z","snapshot_observed_at":"2026-07-06T16:15:00.517258Z","submitted_at":"2023-09-06T04:37:20Z","title":"Certifying LLM Safety against Adversarial Prompting","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2309.02705","snapshot_observed_at":"2026-08-11T12:57:06.202073Z","title":"Certifying LLM safety against adversarial prompting","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2412.13705","last_updated":"2024-12-18T10:49:41Z","snapshot_observed_at":"2026-08-12T09:36:26.801343Z","submitted_at":"2024-12-18T10:49:41Z","title":"Mitigating Adversarial Attacks in LLMs through Defensive Suffix Generation","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-11T12:57:06.202073Z"},"links":{"cited_paper":"/paper/2309.02705","citing_paper":"/paper/2412.13705"},"observation_digest":"sha256:1bd2d35ed748e907c279f2cda8bdc70c7e388dcdf5c3472bce99085c3922f219","observation_id":"b9687f5e-8ce3-42c1-9d87-e8cecd757124","resolution":{"observed_at":"2026-08-11T12:57:06.202073Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2004.01970","last_updated":"2020-10-08T00:41:43Z","snapshot_observed_at":"2026-08-08T14:00:19.815816Z","submitted_at":"2020-04-04T16:25:48Z","title":"BAE: BERT-based Adversarial Examples for Text Classification","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2004.01970","snapshot_observed_at":"2026-08-11T12:57:06.207171Z","title":"BAE: BERT-based adversarial examples for text classification","venue":null,"work_id":null,"year":2004},"citing_paper":{"arxiv_id":"2412.13705","last_updated":"2024-12-18T10:49:41Z","snapshot_observed_at":"2026-08-12T09:36:26.801343Z","submitted_at":"2024-12-18T10:49:41Z","title":"Mitigating Adversarial Attacks in LLMs through Defensive Suffix Generation","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-11T12:57:06.207171Z"},"links":{"cited_paper":"/paper/2004.01970","citing_paper":"/paper/2412.13705"},"observation_digest":"sha256:440c413eef0ed06718bca4c4adbcff631f21254cb5c14c23adc3ed75438e41ca","observation_id":"b96d59f3-4ef7-42ae-a507-3112b88d60d9","resolution":{"observed_at":"2026-08-11T12:57:06.207171Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2402.12959","last_updated":"2024-02-20T12:25:26Z","snapshot_observed_at":"2026-08-12T09:36:33.944049Z","submitted_at":"2024-02-20T12:25:26Z","title":"Prompt Stealing Attacks Against Large Language Models","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2402.12959","snapshot_observed_at":"2026-08-11T12:57:06.212588Z","title":"Prompt stealing attacks against large language models.arXiv preprint arXiv:2402.12959,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2412.13705","last_updated":"2024-12-18T10:49:41Z","snapshot_observed_at":"2026-08-12T09:36:26.801343Z","submitted_at":"2024-12-18T10:49:41Z","title":"Mitigating Adversarial Attacks in LLMs through Defensive Suffix Generation","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-11T12:57:06.212588Z"},"links":{"cited_paper":"/paper/2402.12959","citing_paper":"/paper/2412.13705"},"observation_digest":"sha256:ca40818a2426f71707e11f2b6d74bdbf353454f1eb7c9c5077c53fd9ad206874","observation_id":"4078866e-aba9-4a0f-b8bf-2e96c51bede3","resolution":{"observed_at":"2026-08-11T12:57:06.212588Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2307.15043","last_updated":"2023-12-20T20:48:57Z","snapshot_observed_at":"2026-08-12T09:06:50.363435Z","submitted_at":"2023-07-27T17:49:12Z","title":"Universal and Transferable Adversarial Attacks on Aligned Language Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2307.15043","snapshot_observed_at":"2026-08-11T12:57:06.217313Z","title":"Universal and transferable adversarial attacks on aligned language models","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2412.13705","last_updated":"2024-12-18T10:49:41Z","snapshot_observed_at":"2026-08-12T09:36:26.801343Z","submitted_at":"2024-12-18T10:49:41Z","title":"Mitigating Adversarial Attacks in LLMs through Defensive Suffix Generation","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-11T12:57:06.217313Z"},"links":{"cited_paper":"/paper/2307.15043","citing_paper":"/paper/2412.13705"},"observation_digest":"sha256:0012b1063a2b4b87194581176b9438c948eb172aa489870020275d504842bc34","observation_id":"85e3ca18-1eb1-4d03-871d-965b5cde0181","resolution":{"observed_at":"2026-08-11T12:57:06.217313Z","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":"2026-08-11T12:57:06.536154Z","title":"A closer look at adversarial suffix learning for jailbreaking LLMs","venue":null,"work_id":"b558f8c7-9785-4062-8f65-7d485ac03ff0","year":2024},"citing_paper":{"arxiv_id":"2412.13705","last_updated":"2024-12-18T10:49:41Z","snapshot_observed_at":"2026-08-12T09:36:26.801343Z","submitted_at":"2024-12-18T10:49:41Z","title":"Mitigating Adversarial Attacks in LLMs through Defensive Suffix Generation","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-11T12:57:06.222474Z"},"links":{"citing_paper":"/paper/2412.13705"},"observation_digest":"sha256:d17fed267e8cc0c3b6edc93c08ae64a355fadbd9ac386cf1b5abb63afaf60ee8","observation_id":"fe4dbcd4-c924-49ef-861d-03c4582ed087","resolution":{"observed_at":"2026-08-11T12:57:06.541273Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-12T06:34:41.77262+00:00","source":"crossref"},{"observed_at":"2026-08-12T06:34:36.333875+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2403.08295","last_updated":"2024-04-16T12:52:47Z","snapshot_observed_at":"2026-08-03T03:29:01.959523Z","submitted_at":"2024-03-13T06:59:16Z","title":"Gemma: Open Models Based on Gemini Research and Technology","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2403.08295","snapshot_observed_at":"2026-08-11T12:57:06.227147Z","title":"Llama3.2-1B Model","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2412.13705","last_updated":"2024-12-18T10:49:41Z","snapshot_observed_at":"2026-08-12T09:36:26.801343Z","submitted_at":"2024-12-18T10:49:41Z","title":"Mitigating Adversarial Attacks in LLMs through Defensive Suffix Generation","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-11T12:57:06.227147Z"},"links":{"cited_paper":"/paper/2403.08295","citing_paper":"/paper/2412.13705"},"observation_digest":"sha256:20db9a09369497d6221215931216290b608c3b4bd842434aa502cde2db7f4588","observation_id":"01064f47-da56-47e6-9fcc-d4154fdffc1b","resolution":{"observed_at":"2026-08-11T12:57:06.227147Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2310.06825","last_updated":"2023-10-10T17:54:58Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","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":"2026-08-11T12:57:06.232212Z","title":"Mistral 7B","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2412.13705","last_updated":"2024-12-18T10:49:41Z","snapshot_observed_at":"2026-08-12T09:36:26.801343Z","submitted_at":"2024-12-18T10:49:41Z","title":"Mitigating Adversarial Attacks in LLMs through Defensive Suffix Generation","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-11T12:57:06.232212Z"},"links":{"cited_paper":"/paper/2310.06825","citing_paper":"/paper/2412.13705"},"observation_digest":"sha256:a9e28029e79ac8ecc50775b23085e509a32135bd1c94bd51f4290d5f90a6f3af","observation_id":"2e4efc74-44ab-4342-8bac-d1e852c64037","resolution":{"observed_at":"2026-08-11T12:57:06.232212Z","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":"2026-08-11T12:57:06.237125Z","title":"Llama 2: Open foundation and fine-tuned chat models","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2412.13705","last_updated":"2024-12-18T10:49:41Z","snapshot_observed_at":"2026-08-12T09:36:26.801343Z","submitted_at":"2024-12-18T10:49:41Z","title":"Mitigating Adversarial Attacks in LLMs through Defensive Suffix Generation","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-11T12:57:06.237125Z"},"links":{"cited_paper":"/paper/2307.09288","citing_paper":"/paper/2412.13705"},"observation_digest":"sha256:214dce7465e7a217890b23c706aabfbee03ae7e7f6e73557d8e0cbb9e53acfb7","observation_id":"cdb648a9-50a4-4cd1-bc55-d6c518615a3f","resolution":{"observed_at":"2026-08-11T12:57:06.237125Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2109.07958","last_updated":"2022-05-08T02:43:02Z","snapshot_observed_at":"2026-08-03T16:26:48.747700Z","submitted_at":"2021-09-08T17:15:27Z","title":"TruthfulQA: Measuring How Models Mimic Human Falsehoods","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2109.07958","snapshot_observed_at":"2026-08-11T12:57:06.242039Z","title":"Llama2-13B Model","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2412.13705","last_updated":"2024-12-18T10:49:41Z","snapshot_observed_at":"2026-08-12T09:36:26.801343Z","submitted_at":"2024-12-18T10:49:41Z","title":"Mitigating Adversarial Attacks in LLMs through Defensive Suffix Generation","version":1},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-11T12:57:06.242039Z"},"links":{"cited_paper":"/paper/2109.07958","citing_paper":"/paper/2412.13705"},"observation_digest":"sha256:d5a6d68c99be465c1d93666d996bdbcc6ccfedddf495e45c6c1f81e9b2539ff1","observation_id":"69e3529a-6040-474a-821d-0dd0412edb6a","resolution":{"observed_at":"2026-08-11T12:57:06.242039Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2310.00905","last_updated":"2024-06-20T14:15:23Z","snapshot_observed_at":"2026-08-12T19:52:16.624565Z","submitted_at":"2023-10-02T05:23:34Z","title":"All Languages Matter: On the Multilingual Safety of Large Language Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2310.00905","snapshot_observed_at":"2026-08-11T12:57:06.186601Z","title":"All Languages Matter: On the Multilingual Safety of Large Language Models","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2412.13705","last_updated":"2024-12-18T10:49:41Z","snapshot_observed_at":"2026-08-12T09:36:26.801343Z","submitted_at":"2024-12-18T10:49:41Z","title":"Mitigating Adversarial Attacks in LLMs through Defensive Suffix Generation","version":1},"reference_index":2020,"source":"pdf_text","source_observed_at":"2026-08-11T12:57:06.186601Z"},"links":{"cited_paper":"/paper/2310.00905","citing_paper":"/paper/2412.13705"},"observation_digest":"sha256:ba2c9823f7d64ab7ee093b27354362cd929c31cc60bd2715738430e519a5cac8","observation_id":"a8fd48f0-1b5d-4283-b2e9-8b3c6f13f295","resolution":{"observed_at":"2026-08-11T12:57:06.186601Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1904.09675","last_updated":"2020-02-24T18:59:28Z","snapshot_observed_at":"2026-07-29T15:42:51.774083Z","submitted_at":"2019-04-21T23:08:53Z","title":"BERTScore: Evaluating Text Generation with BERT","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1904.09675","snapshot_observed_at":"2026-08-11T12:57:06.247853Z","title":"BERTScore: Evaluating text generation with BERT","venue":null,"work_id":null,"year":1904},"citing_paper":{"arxiv_id":"2412.13705","last_updated":"2024-12-18T10:49:41Z","snapshot_observed_at":"2026-08-12T09:36:26.801343Z","submitted_at":"2024-12-18T10:49:41Z","title":"Mitigating Adversarial Attacks in LLMs through Defensive Suffix Generation","version":1},"reference_index":2021,"source":"pdf_text","source_observed_at":"2026-08-11T12:57:06.247853Z"},"links":{"cited_paper":"/paper/1904.09675","citing_paper":"/paper/2412.13705"},"observation_digest":"sha256:c70b5af578cf858a6f4a6c67fd1000af19788086824425cf84846daec8f67da4","observation_id":"7b6edafa-6db6-4535-92a5-cc31c4735d63","resolution":{"observed_at":"2026-08-11T12:57:06.247853Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2309.07755","last_updated":"2023-09-14T14:41:46Z","snapshot_observed_at":"2026-08-12T09:36:33.919963Z","submitted_at":"2023-09-14T14:41:46Z","title":"Generative AI Text Classification using Ensemble LLM Approaches","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2309.07755","snapshot_observed_at":"2026-08-11T12:57:06.159703Z","title":"Generative AI text classification using ensemble LLM approaches","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2412.13705","last_updated":"2024-12-18T10:49:41Z","snapshot_observed_at":"2026-08-12T09:36:26.801343Z","submitted_at":"2024-12-18T10:49:41Z","title":"Mitigating Adversarial Attacks in LLMs through Defensive Suffix Generation","version":1},"reference_index":2022,"source":"pdf_text","source_observed_at":"2026-08-11T12:57:06.159703Z"},"links":{"cited_paper":"/paper/2309.07755","citing_paper":"/paper/2412.13705"},"observation_digest":"sha256:62d7b988d5e0ef8bf1301eabaa294e474368ae4ea62e7af98992f74791872ba5","observation_id":"793f5185-0344-40be-a117-8a0723f518d9","resolution":{"observed_at":"2026-08-11T12:57:06.159703Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2306.04387","last_updated":"2023-06-08T13:44:24Z","snapshot_observed_at":"2026-08-12T09:36:48.705293Z","submitted_at":"2023-06-07T12:35:37Z","title":"M$^3$IT: A Large-Scale Dataset towards Multi-Modal Multilingual Instruction Tuning","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2306.04387","snapshot_observed_at":"2026-08-11T12:57:06.165383Z","title":"M3IT: A Large-Scale Dataset towards Multi-Modal Multilingual Instruction Tuning","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2412.13705","last_updated":"2024-12-18T10:49:41Z","snapshot_observed_at":"2026-08-12T09:36:26.801343Z","submitted_at":"2024-12-18T10:49:41Z","title":"Mitigating Adversarial Attacks in LLMs through Defensive Suffix Generation","version":1},"reference_index":2023,"source":"pdf_text","source_observed_at":"2026-08-11T12:57:06.165383Z"},"links":{"cited_paper":"/paper/2306.04387","citing_paper":"/paper/2412.13705"},"observation_digest":"sha256:a7554df9baccbda246bd0cb84da50b94c15e1040d9e5c96b798c8a7c7b2292ef","observation_id":"2db8fa1d-12b9-43c0-b146-e254233ba213","resolution":{"observed_at":"2026-08-11T12:57:06.165383Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2310.13345","last_updated":"2023-10-20T08:16:46Z","snapshot_observed_at":"2026-08-12T09:36:33.300513Z","submitted_at":"2023-10-20T08:16:46Z","title":"An LLM can Fool Itself: A Prompt-Based Adversarial Attack","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2310.13345","snapshot_observed_at":"2026-08-11T12:57:06.175752Z","title":"An LLM can Fool Itself: A Prompt-Based Adversarial Attack","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2412.13705","last_updated":"2024-12-18T10:49:41Z","snapshot_observed_at":"2026-08-12T09:36:26.801343Z","submitted_at":"2024-12-18T10:49:41Z","title":"Mitigating Adversarial Attacks in LLMs through Defensive Suffix Generation","version":1},"reference_index":2024,"source":"pdf_text","source_observed_at":"2026-08-11T12:57:06.175752Z"},"links":{"cited_paper":"/paper/2310.13345","citing_paper":"/paper/2412.13705"},"observation_digest":"sha256:487e1b43721977e5deb912d08e456a0bb5c40700303feec303e89727d402c237","observation_id":"b5a25a57-915c-457e-8355-faca9fa07997","resolution":{"observed_at":"2026-08-11T12:57:06.175752Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"paper":{"arxiv_id":"2412.13705","last_updated":"2024-12-18T10:49:41Z","latest_version":1,"primary_category":"cs.CV","snapshot_observed_at":"2026-08-12T09:36:26.801343Z","submitted_at":"2024-12-18T10:49:41Z","title":"Mitigating Adversarial Attacks in LLMs through Defensive Suffix Generation"},"reference_resolution":{"displayed":18,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":15,"verified_exact":1,"verified_fuzzy":2},"total_outbound_references":18},"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":"2026-08-12T06:34:41.77262+00:00","source":"crossref"},{"observed_at":"2026-08-12T06:34:36.333875+00:00","source":"retraction_watch"}],"thesis":"As of 12 August 2026, this Paper Citation Record lists 18 of 18 outbound references and 1 inbound Pith citation observation for arXiv:2412.13705."}