{"as_of":"2026-08-09T18:59:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:56daf2e41a44cd9f057b705d2f394dfdafe48ce5616b17f724d6198ba7bc1c17","coverage":[{"denominator":0,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":0,"source":"paper_references, paper_reference_links","source_observed_at":null,"state":"measured"},{"denominator":9,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":9,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-09T06:31:02.800959+00:00","state":"measured"},{"denominator":9,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":9,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-07T14:15:14.184417Z","state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"arxiv_reference","source_observed_at":"2026-07-03T06:17:42.479409Z","state":"measured"}],"external_citation_measurements":[],"inbound":[{"citation":{"cited_paper":{"arxiv_id":"2412.18826","last_updated":"2024-12-25T08:31:53Z","snapshot_observed_at":"2026-07-06T20:13:03.722546Z","submitted_at":"2024-12-25T08:31:53Z","title":"RapGuard: Safeguarding Multimodal Large Language Models via Rationale-aware Defensive Prompting","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2412.18826","snapshot_observed_at":"2026-08-07T14:13:51.539892Z","title":"Rapguard: Safeguarding multimodal large language models via rationale- aware defensive prompting,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2505.19690","last_updated":"2025-05-26T08:49:19Z","snapshot_observed_at":"2026-08-08T11:27:57.969115Z","submitted_at":"2025-05-26T08:49:19Z","title":"Beyond Safe Answers: A Benchmark for Evaluating True Risk Awareness in Large Reasoning Models","version":1},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-07T14:13:51.539892Z"},"links":{"cited_paper":"/paper/2412.18826","citing_paper":"/paper/2505.19690"},"observation_digest":"sha256:33b9b6c3808e1d021215efae107c89749c1de01b87fa912340ffa2861e09d0ac","observation_id":"7277bf91-625e-450a-a6b9-dc573a7ecbe7","resolution":{"observed_at":"2026-08-07T14:13:51.539892Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2412.18826","last_updated":"2024-12-25T08:31:53Z","snapshot_observed_at":"2026-07-06T20:13:03.722546Z","submitted_at":"2024-12-25T08:31:53Z","title":"RapGuard: Safeguarding Multimodal Large Language Models via Rationale-aware Defensive Prompting","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2412.18826","snapshot_observed_at":"2026-08-07T13:35:58.091606Z","title":"Rapguard: Safeguarding multimodal large language models via rationale-aware defensive prompting, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2505.21327","last_updated":"2025-05-27T15:23:23Z","snapshot_observed_at":"2026-08-08T22:42:35.699276Z","submitted_at":"2025-05-27T15:23:23Z","title":"MME-Reasoning: A Comprehensive Benchmark for Logical Reasoning in MLLMs","version":1},"reference_index":32,"source":"arxiv_source","source_observed_at":"2026-08-07T13:35:58.091606Z"},"links":{"cited_paper":"/paper/2412.18826","citing_paper":"/paper/2505.21327"},"observation_digest":"sha256:7d5a3c01cd1e6fb0c9f0f11bf13aca8fcd3eea92c0a9b7dd7baffec6d30bd316","observation_id":"06fbccd6-ccf5-4b53-a32f-ac62a3b3c0c9","resolution":{"observed_at":"2026-08-07T13:35:58.091606Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2412.18826","last_updated":"2024-12-25T08:31:53Z","snapshot_observed_at":"2026-07-06T20:13:03.722546Z","submitted_at":"2024-12-25T08:31:53Z","title":"RapGuard: Safeguarding Multimodal Large Language Models via Rationale-aware Defensive Prompting","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2412.18826","snapshot_observed_at":"2026-08-07T14:15:14.184417Z","title":"RapGuard: Safeguarding multimodal large language models via rationale-aware defensive prompting,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2505.23793","last_updated":"2025-05-26T08:39:14Z","snapshot_observed_at":"2026-08-08T01:18:19.713600Z","submitted_at":"2025-05-26T08:39:14Z","title":"USB: A Comprehensive and Unified Safety Evaluation Benchmark for Multimodal Large Language Models","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-07T14:15:14.184417Z"},"links":{"cited_paper":"/paper/2412.18826","citing_paper":"/paper/2505.23793"},"observation_digest":"sha256:1b7438812a89fe56d79f58315915545c0ae98330e1e0f530d58032fd90d19c8c","observation_id":"a41746f6-c695-405c-87a4-f2817a9eaf7f","resolution":{"observed_at":"2026-08-07T14:15:14.184417Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2412.18826","last_updated":"2024-12-25T08:31:53Z","snapshot_observed_at":"2026-07-06T20:13:03.722546Z","submitted_at":"2024-12-25T08:31:53Z","title":"RapGuard: Safeguarding Multimodal Large Language Models via Rationale-aware Defensive Prompting","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2412.18826","snapshot_observed_at":"2026-08-05T21:18:42.356913Z","title":null,"venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2508.09016","last_updated":"2025-09-10T05:08:47Z","snapshot_observed_at":"2026-08-09T06:48:43.608452Z","submitted_at":"2025-08-12T15:30:44Z","title":"A Survey on Training-free Alignment of Large Language Models","version":4},"reference_index":41,"source":"arxiv_source","source_observed_at":"2026-08-05T21:18:42.356913Z"},"links":{"cited_paper":"/paper/2412.18826","citing_paper":"/paper/2508.09016"},"observation_digest":"sha256:22ca4ef4bb8f974e9d6189aed417abaad1258d2cdd5d7f0db08a11d84179106f","observation_id":"a2bf540d-4aa7-4f01-ac6a-db15f91a4cdf","resolution":{"observed_at":"2026-08-05T21:18:42.356913Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2412.18826","last_updated":"2024-12-25T08:31:53Z","snapshot_observed_at":"2026-07-06T20:13:03.722546Z","submitted_at":"2024-12-25T08:31:53Z","title":"RapGuard: Safeguarding Multimodal Large Language Models via Rationale-aware Defensive Prompting","version":1},"cited_work":{"arxiv_id":"2412.18826","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2412.18826","snapshot_observed_at":"2026-07-03T06:17:42.479409Z","title":"Rap- guard: Safeguarding multimodal large language models via rationale-aware defensive prompting","venue":null,"work_id":"07a6dc61-c076-4dba-a3fb-5845238d9f2f","year":2024},"citing_paper":{"arxiv_id":"2604.08846","last_updated":"2026-04-10T01:01:56Z","snapshot_observed_at":"2026-08-02T14:54:38.934183Z","submitted_at":"2026-04-10T01:01:56Z","title":"Dictionary-Aligned Concept Control for Safeguarding Multimodal LLMs","version":1},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-05-10T18:04:05.157103Z"},"links":{"cited_paper":"/paper/2412.18826","citing_paper":"/paper/2604.08846"},"observation_digest":"sha256:e7239da3bf2371a1e441ad168efa9a58f60abb65d19b1c8fb6f8e16a615af3e8","observation_id":"6839cb4a-4f00-4393-a924-a6906d3e6cd1","resolution":{"observed_at":"2026-05-11T05:35:57.229665Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2412.18826","last_updated":"2024-12-25T08:31:53Z","snapshot_observed_at":"2026-07-06T20:13:03.722546Z","submitted_at":"2024-12-25T08:31:53Z","title":"RapGuard: Safeguarding Multimodal Large Language Models via Rationale-aware Defensive Prompting","version":1},"cited_work":{"arxiv_id":"2412.18826","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2412.18826","snapshot_observed_at":"2026-07-03T06:17:42.479409Z","title":"Rap- guard: Safeguarding multimodal large language models via rationale-aware defensive prompting","venue":null,"work_id":"07a6dc61-c076-4dba-a3fb-5845238d9f2f","year":2024},"citing_paper":{"arxiv_id":"2606.10904","last_updated":"2026-07-16T08:23:49Z","snapshot_observed_at":"2026-08-05T14:18:44.309610Z","submitted_at":"2026-06-09T14:13:54Z","title":"Auditing Inference-Time Defense Evaluation for Multimodal Large Language Models","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-06-27T12:44:15.097414Z"},"links":{"cited_paper":"/paper/2412.18826","citing_paper":"/paper/2606.10904"},"observation_digest":"sha256:5edbff8b8e75d32bd65650b0bfef1e44e820a133a8194665b1887c2f82c822b4","observation_id":"62fed770-48cb-4f0e-bafa-7a6ca09f30f6","resolution":{"observed_at":"2026-07-03T06:17:42.480736Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2412.18826","last_updated":"2024-12-25T08:31:53Z","snapshot_observed_at":"2026-07-06T20:13:03.722546Z","submitted_at":"2024-12-25T08:31:53Z","title":"RapGuard: Safeguarding Multimodal Large Language Models via Rationale-aware Defensive Prompting","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2412.18826","snapshot_observed_at":"2026-07-15T10:53:29.444919Z","title":"arXiv preprint arXiv:2412.18826 (2024)","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2606.10904","last_updated":"2026-07-16T08:23:49Z","snapshot_observed_at":"2026-08-05T14:18:44.309610Z","submitted_at":"2026-06-09T14:13:54Z","title":"Auditing Inference-Time Defense Evaluation for Multimodal Large Language Models","version":2},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-07-15T10:53:29.444919Z"},"links":{"cited_paper":"/paper/2412.18826","citing_paper":"/paper/2606.10904"},"observation_digest":"sha256:e7cc8e2db06dd3de44243978be0d575552ea56b030f06b9e9c500f8aff07d951","observation_id":"6182b365-3376-46dd-b98d-d28d04dbe050","resolution":{"observed_at":"2026-07-15T10:53:29.444919Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2412.18826","last_updated":"2024-12-25T08:31:53Z","snapshot_observed_at":"2026-07-06T20:13:03.722546Z","submitted_at":"2024-12-25T08:31:53Z","title":"RapGuard: Safeguarding Multimodal Large Language Models via Rationale-aware Defensive Prompting","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2412.18826","snapshot_observed_at":"2026-08-02T11:56:59.624836Z","title":"arXiv preprint arXiv:2412.18826 (2024)","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2606.10904","last_updated":"2026-07-16T08:23:49Z","snapshot_observed_at":"2026-08-05T14:18:44.309610Z","submitted_at":"2026-06-09T14:13:54Z","title":"Auditing Inference-Time Defense Evaluation for Multimodal Large Language Models","version":3},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-02T11:56:59.624836Z"},"links":{"cited_paper":"/paper/2412.18826","citing_paper":"/paper/2606.10904"},"observation_digest":"sha256:84a32b737f948ea20990e43391618f3a381fb29686276a2b1bf0c91138ec5b36","observation_id":"609026f8-ace2-42ce-ad81-a2ce0c9bcd93","resolution":{"observed_at":"2026-08-02T11:56:59.624836Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2412.18826","last_updated":"2024-12-25T08:31:53Z","snapshot_observed_at":"2026-07-06T20:13:03.722546Z","submitted_at":"2024-12-25T08:31:53Z","title":"RapGuard: Safeguarding Multimodal Large Language Models via Rationale-aware Defensive Prompting","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2412.18826","snapshot_observed_at":"2026-07-14T17:33:49.041193Z","title":"arXiv preprint arXiv:2412.18826 (2024)","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.09697","last_updated":"2026-06-20T12:09:13Z","snapshot_observed_at":"2026-08-08T10:05:10.130285Z","submitted_at":"2026-06-20T12:09:13Z","title":"Safe responses matter: Output-aware safety guardrail mitigate over-refusal in MLLMs","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-07-14T17:33:49.041193Z"},"links":{"cited_paper":"/paper/2412.18826","citing_paper":"/paper/2607.09697"},"observation_digest":"sha256:511d50189048b000e21be101a361cc2e376c35ddc1a795ede5bc9b73b564501c","observation_id":"09cb3e5f-c354-47c1-89b1-14d8f3458f5e","resolution":{"observed_at":"2026-07-14T17:33:49.041193Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"links":{"evidence":"/evidence","html":"/paper/2412.18826/citation-record","integrity":"/paper/2412.18826/integrity","json":"/paper/2412.18826/citation-record.json","paper":"/paper/2412.18826"},"outbound":[],"paper":{"arxiv_id":"2412.18826","last_updated":"2024-12-25T08:31:53Z","latest_version":1,"primary_category":"cs.CL","snapshot_observed_at":"2026-07-06T20:13:03.722546Z","submitted_at":"2024-12-25T08:31:53Z","title":"RapGuard: Safeguarding Multimodal Large Language Models via Rationale-aware Defensive Prompting"},"reference_resolution":{"displayed":0,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":0,"verified_exact":0,"verified_fuzzy":0},"total_outbound_references":0},"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":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+00:00","source":"retraction_watch"}],"thesis":"As of 9 August 2026, this Paper Citation Record lists 0 of 0 outbound references and 9 inbound Pith citation observations for arXiv:2412.18826."}