{"as_of":"2026-08-24T03:41:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:8e83835cfe7f7ae2d40387fcfcb35bd9e15d3dd713124b63e9aad556bd614623","coverage":[{"denominator":74,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":74,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-02T07:06:24.001849Z","state":"measured"},{"denominator":74,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":74,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-23T06:30:58.430688+00:00","state":"measured"},{"denominator":0,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":0,"source":"paper_references, paper_reference_links","source_observed_at":null,"state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"cited_works","source_observed_at":null,"state":"measured"}],"external_citation_measurements":[],"inbound":[],"links":{"evidence":"/evidence","html":"/paper/2607.13078/citation-record","integrity":"/paper/2607.13078/integrity","json":"/paper/2607.13078/citation-record.json","paper":"/paper/2607.13078"},"outbound":[{"citation":{"cited_paper":{"arxiv_id":"2512.13040","last_updated":"2026-04-08T20:16:25Z","snapshot_observed_at":"2026-08-11T12:29:28.227636Z","submitted_at":"2025-12-15T07:09:11Z","title":"Understanding Structured Financial Data with LLMs: A Case Study on Fraud Detection","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2512.13040","snapshot_observed_at":"2026-08-02T07:06:17.476603Z","title":"Understanding structured financial data with LLMs: A case study on fraud detection, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:17.476603Z"},"links":{"cited_paper":"/paper/2512.13040","citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:3474c7bbd2b9b44f68ee2df0d55af8961ee6514006b4928a52a3b489bde7a66e","observation_id":"2378bb77-cbef-49bc-bcec-e9818a8fc263","resolution":{"observed_at":"2026-08-02T07:06:17.476603Z","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":"2026-08-02T07:06:17.582243Z","title":"FLAG: Fraud detection with LLM-enhanced graph neural network","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:17.582243Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:f05ae7a37f2ccce3b01b597290dfc9c70c833fae3ba06c31f8e0709c85d403f2","observation_id":"63962b07-74e2-4e9d-b827-c35242452fe1","resolution":{"observed_at":"2026-08-02T07:06:17.582243Z","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":"2026-08-02T07:06:17.760024Z","title":"LLM-enhanced self-evolving reinforcement learning for multi-step e-commerce pay- ment fraud risk detection","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:17.760024Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:f26c0fb299b30f6af8deef9eb44fb04c4262ef237cb352568a6136709dcfbb21","observation_id":"df8422c3-852f-4496-ad3e-2d40102aa1d9","resolution":{"observed_at":"2026-08-02T07:06:17.760024Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2501.15290","last_updated":"2025-01-25T17:58:05Z","snapshot_observed_at":"2026-08-20T04:06:58.677689Z","submitted_at":"2025-01-25T17:58:05Z","title":"Advanced Real-Time Fraud Detection Using RAG-Based LLMs","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2501.15290","snapshot_observed_at":"2026-08-02T07:06:17.938720Z","title":"Advanced real-time fraud detection using RAG-based LLMs, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:17.938720Z"},"links":{"cited_paper":"/paper/2501.15290","citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:421851180c4846b36d917e2266f4f2253b23f343d63922a84356e7d87c43a897","observation_id":"d3c6077c-0cb0-43a4-ba4d-3d203bc62f9f","resolution":{"observed_at":"2026-08-02T07:06:17.938720Z","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":"2026-08-02T07:06:18.063127Z","title":"LLM-assistedauthenticationandfrauddetection,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:18.063127Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:3ad10be56c47b136d1184e9fddf9b2c181fd8a1131348645f7aa1194e1d981c8","observation_id":"dd79c548-c47b-4860-b18a-a5493c27b52c","resolution":{"observed_at":"2026-08-02T07:06:18.063127Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2404.05993","last_updated":"2024-09-11T14:42:29Z","snapshot_observed_at":"2026-08-22T03:42:00.219266Z","submitted_at":"2024-04-09T03:54:28Z","title":"AEGIS: Online Adaptive AI Content Safety Moderation with Ensemble of LLM Experts","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2404.05993","snapshot_observed_at":"2026-08-02T07:06:18.415677Z","title":"AEGIS: Online adaptive AI content safety moderation with ensemble of LLM experts, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:18.415677Z"},"links":{"cited_paper":"/paper/2404.05993","citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:07d9e7a60eeef739ed1cac43b78ed730df26dfed6a790a9fabc8624ea3a22254","observation_id":"7171ac90-634b-41f6-ae0b-ab4cbebbfe30","resolution":{"observed_at":"2026-08-02T07:06:18.415677Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2410.13155","last_updated":"2025-02-10T02:02:11Z","snapshot_observed_at":"2026-08-16T13:08:35.523123Z","submitted_at":"2024-10-17T02:16:37Z","title":"SLM-Mod: Small Language Models Surpass LLMs at Content Moderation","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2410.13155","snapshot_observed_at":"2026-08-02T07:06:18.561141Z","title":"SLM-Mod: Small language models surpass LLMs at content moderation","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:18.561141Z"},"links":{"cited_paper":"/paper/2410.13155","citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:950389ab216c12cc9b4a9988b93312026b05d4d7666b95bd7f47e288e8b0233f","observation_id":"562f4291-508b-4f6b-bdd2-96806c3837a2","resolution":{"observed_at":"2026-08-02T07:06:18.561141Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2508.19932","last_updated":"2026-05-01T07:50:38Z","snapshot_observed_at":"2026-08-07T17:04:40.935858Z","submitted_at":"2025-08-27T14:47:33Z","title":"CASE: An Agentic AI Framework for Enhancing Scam Intelligence in Digital Payments","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2508.19932","snapshot_observed_at":"2026-08-02T07:06:18.728116Z","title":"Agentic AI framework for enhancing scam intelligence in digital payments, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:18.728116Z"},"links":{"cited_paper":"/paper/2508.19932","citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:15068650848d527e97bf16c50301276596a2f22481f637606fa666d8b1ce36d5","observation_id":"e2777191-b6e2-40aa-a1f6-ef7c773a282d","resolution":{"observed_at":"2026-08-02T07:06:18.728116Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2503.20796","last_updated":"2025-03-22T23:37:35Z","snapshot_observed_at":"2026-08-18T16:26:06.688492Z","submitted_at":"2025-03-22T23:37:35Z","title":"EXPLICATE: Enhancing Phishing Detection through Explainable AI and LLM-Powered Interpretability","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2503.20796","snapshot_observed_at":"2026-08-02T07:06:18.864784Z","title":"EXPLICATE: Enhancing phishing detection through explainable AI and LLM-powered interpretability, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:18.864784Z"},"links":{"cited_paper":"/paper/2503.20796","citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:18ce4a9d5713d2e2ab372d0039b0165ceac8ea0943526039207443556833dbe4","observation_id":"692958b1-aeb4-41f3-b15e-49334dd31303","resolution":{"observed_at":"2026-08-02T07:06:18.864784Z","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":"2026-08-02T07:06:18.994217Z","title":"Co-investigator AI: The rise of agentic AI for smarter, trustworthy AML compliance narratives, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:18.994217Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:c1072dba1bae640c64309db115b248290dc6c50510c27b09f0ac6bf9a1dd8755","observation_id":"dac0124c-1020-4629-ae97-f426f0b615fa","resolution":{"observed_at":"2026-08-02T07:06:18.994217Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2506.11635","last_updated":"2025-06-13T10:05:43Z","snapshot_observed_at":"2026-08-21T10:11:14.221922Z","submitted_at":"2025-06-13T10:05:43Z","title":"FAA Framework: A Large Language Model-Based Approach for Credit Card Fraud Investigations","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2506.11635","snapshot_observed_at":"2026-08-02T07:06:19.069283Z","title":"FAA framework: A large language model-based approach for credit card fraud investigations, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:19.069283Z"},"links":{"cited_paper":"/paper/2506.11635","citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:546864f9183c0d4721bdf7dbbe17be84b0254446ca51e3cdffcd4f11d609c62c","observation_id":"0808aa5f-557e-464c-90f2-40e3f6041d0a","resolution":{"observed_at":"2026-08-02T07:06:19.069283Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2410.02644","last_updated":"2025-05-30T03:50:33Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2024-10-03T16:30:47Z","title":"Agent Security Bench (ASB): Formalizing and Benchmarking Attacks and Defenses in LLM-based Agents","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2410.02644","snapshot_observed_at":"2026-08-02T07:06:19.120982Z","title":"Agent security bench (ASB): Formalizing and bench- marking attacks and defenses in LLM-based agents, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:19.120982Z"},"links":{"cited_paper":"/paper/2410.02644","citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:1453e3afccc16facd62f09afac62f944675b2e0091356eedef725575560282f0","observation_id":"8ecaf8ac-cb16-46f9-b9d3-1c1419778c20","resolution":{"observed_at":"2026-08-02T07:06:19.120982Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2402.07867","last_updated":"2024-08-13T01:55:06Z","snapshot_observed_at":"2026-08-18T12:32:21.937066Z","submitted_at":"2024-02-12T18:28:36Z","title":"PoisonedRAG: Knowledge Corruption Attacks to Retrieval-Augmented Generation of Large Language Models","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2402.07867","snapshot_observed_at":"2026-08-02T07:06:19.177718Z","title":"PoisonedRAG: Knowledge cor- ruption attacks to retrieval-augmented generation of large language models, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:19.177718Z"},"links":{"cited_paper":"/paper/2402.07867","citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:74184209f14e6541de610414de14c5e6a46620445360f4955d307bd0dc45d008","observation_id":"ee90f312-5107-411e-a5a3-e2e13e65d86b","resolution":{"observed_at":"2026-08-02T07:06:19.177718Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2512.12844","last_updated":"2026-04-27T02:26:44Z","snapshot_observed_at":"2026-08-17T11:35:06.879112Z","submitted_at":"2025-12-14T21:18:28Z","title":"Selective Conformal Risk Control","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2512.12844","snapshot_observed_at":"2026-08-02T07:06:19.248857Z","title":"Selective conformal risk control, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:19.248857Z"},"links":{"cited_paper":"/paper/2512.12844","citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:40a2d645b5166b15d04b82965da5ce7c02ff05c738361e6e8317fa6a3d373df2","observation_id":"119e0060-e43c-4a9f-910d-c79a89ca5683","resolution":{"observed_at":"2026-08-02T07:06:19.248857Z","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":"2026-08-02T07:06:19.318217Z","title":"Enhancing the interpretability of SHAP values using LLMs, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:19.318217Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:96683a7ad72838484e6ffa645508d9c576d13d0144ecaa465fe4e7f03e397248","observation_id":"0c77aa58-f4ba-4a7d-a9c2-9f84f1761d16","resolution":{"observed_at":"2026-08-02T07:06:19.318217Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2406.02622","last_updated":"2024-06-03T19:27:46Z","snapshot_observed_at":"2026-08-19T03:06:14.133584Z","submitted_at":"2024-06-03T19:27:46Z","title":"Safeguarding Large Language Models: A Survey","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2406.02622","snapshot_observed_at":"2026-08-02T07:06:19.372134Z","title":"Safeguarding large language models: A survey, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:19.372134Z"},"links":{"cited_paper":"/paper/2406.02622","citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:5691fde6653250b8483a16665abceacd43563844008041d347f28f5fb3a579b6","observation_id":"9945c231-4833-45d2-801a-6200a58d1a14","resolution":{"observed_at":"2026-08-02T07:06:19.372134Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2502.05206","last_updated":"2026-04-14T16:10:41Z","snapshot_observed_at":"2026-08-20T09:36:18.005089Z","submitted_at":"2025-02-02T05:14:22Z","title":"Safety at Scale: A Comprehensive Survey of Large Model and Agent Safety","version":6},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2502.05206","snapshot_observed_at":"2026-08-02T07:06:19.432277Z","title":"Safety at scale: A comprehensive survey of large model and agent safety, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:19.432277Z"},"links":{"cited_paper":"/paper/2502.05206","citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:313b4282da0ceb0f12451b82a2e8afd9225a75068c391a61b9e63f0c8ca1017c","observation_id":"ca918491-74ac-4d57-8dff-e52206643b57","resolution":{"observed_at":"2026-08-02T07:06:19.432277Z","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":"2026-08-02T07:06:19.509653Z","title":"A survey on large language model (LLM) security and privacy: The good, the bad, and the ugly,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:19.509653Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:9b42ecbd31db39514df75014e338ad3c01f8883b617b824985b4df07aef63dcc","observation_id":"b9eda4c6-3cf6-4826-bd79-31809754c8f7","resolution":{"observed_at":"2026-08-02T07:06:19.509653Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2502.00201","last_updated":"2025-07-30T04:32:58Z","snapshot_observed_at":"2026-08-17T20:39:42.026976Z","submitted_at":"2025-01-31T22:31:50Z","title":"Year-over-Year Developments in Financial Fraud Detection via Deep Learning: A Systematic Literature Review","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2502.00201","snapshot_observed_at":"2026-08-02T07:06:19.643424Z","title":"Year-over- year developments in financial fraud detection via deep learning, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:19.643424Z"},"links":{"cited_paper":"/paper/2502.00201","citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:5c42f6cd0c038d3df32cf5a221ef1752ee3dde6a4e30d42eeac33287da616764","observation_id":"e1a0da57-ca2a-44db-9c48-ba7bfcd59edb","resolution":{"observed_at":"2026-08-02T07:06:19.643424Z","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":"2026-08-02T07:06:19.723260Z","title":"Large language models for financial fraud detection: A systematic review of methodologies, performance, and future directions","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:19.723260Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:0d1c98145946f205fce1ce77c4922fff5ff6c38ad1b26194128eff96c1256032","observation_id":"74d2ec6a-8104-4ff0-b204-d393aa1da4b9","resolution":{"observed_at":"2026-08-02T07:06:19.723260Z","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":"10.1186/s40163-025-00248-8","metadata_source":"openalex","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"Applications of AI-based models for online fraud detection and analysis.Crime Science, 14(7), 2025","venue":"Crime Science","work_id":"4b42eb24-1caf-4ea7-b867-db13508e547a","year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:19.782753Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:6325d1bad19e71a8a9a0215117c386d23464e7a7f3b34e10e12dc8a112f5ebb6","observation_id":"3700b3f8-b272-4154-946e-0e9a235d1b49","resolution":{"observed_at":"2026-08-02T07:08:28.670495Z","resolver_source":"doi","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-02T07:06:19.863146Z","title":"Large language models in the abuse detection pipeline, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:19.863146Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:6316bf4ee851694350a9d4bf66ae1f4321d9ad6c07aeb537530aead76e80b4d0","observation_id":"b2426ecc-b60c-4424-afd8-6ac0aecb8abf","resolution":{"observed_at":"2026-08-02T07:06:19.863146Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2602.23636","last_updated":"2026-04-15T12:34:59Z","snapshot_observed_at":"2026-08-14T07:40:06.913771Z","submitted_at":"2026-02-27T03:14:19Z","title":"FlexGuard: Continuous Risk Scoring for Strictness-Adaptive LLM Content Moderation","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2602.23636","snapshot_observed_at":"2026-08-02T07:06:19.932013Z","title":"FlexGuard: Continuous risk scoring for strictness-adaptive LLM content moderation, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:19.932013Z"},"links":{"cited_paper":"/paper/2602.23636","citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:a882e6a5b1d516a8a95075c17a87bffcdd1cf08741232e6058b07ba131ab1c8b","observation_id":"e7c6c49d-d04d-4a18-88c8-2cc5ad795f9b","resolution":{"observed_at":"2026-08-02T07:06:19.932013Z","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":"2026-08-02T07:06:20.000754Z","title":"MeasuringwhatLLMsthinktheydo: SHAPfaithfulnessanddeployability on financial tabular classification, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:20.000754Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:c0cc72a80849f0cd646a53948005eb32547e3454be4dee6b515b24463246f989","observation_id":"c8488e9f-949f-4c99-8c80-3fe41c0255bc","resolution":{"observed_at":"2026-08-02T07:06:20.000754Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2508.05775","last_updated":"2026-07-27T20:50:40Z","snapshot_observed_at":"2026-08-23T20:50:48.930812Z","submitted_at":"2025-08-07T18:42:16Z","title":"Guardians and Offenders: A Survey on Harmful Content Generation and Safety Mitigation of LLM","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2508.05775","snapshot_observed_at":"2026-08-02T07:06:20.077337Z","title":"Guardians and offenders: A survey on harmful content generation and safety mitigation of LLM, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:20.077337Z"},"links":{"cited_paper":"/paper/2508.05775","citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:50f8309ce6c6d2b857733280b95876023921a2c637028ba73c9e94dec39f37e7","observation_id":"62ac8f37-4f52-40ad-b1cc-44d7042493d1","resolution":{"observed_at":"2026-08-02T07:06:20.077337Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2504.17704","last_updated":"2025-05-24T12:50:56Z","snapshot_observed_at":"2026-08-19T22:04:02.585395Z","submitted_at":"2025-04-24T16:11:01Z","title":"Safety in Large Reasoning Models: A Survey","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2504.17704","snapshot_observed_at":"2026-08-02T07:06:20.139263Z","title":"Safety in large reasoning models: A survey, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:20.139263Z"},"links":{"cited_paper":"/paper/2504.17704","citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:3bb8cfa6ba4e5f3b1d0fc5ebb4457460ebef59c7c1c12e4fd94979bb05b64792","observation_id":"61f9dc42-46b5-4970-9d3c-59bc007c3153","resolution":{"observed_at":"2026-08-02T07:06:20.139263Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2507.21653","last_updated":"2025-07-29T10:10:47Z","snapshot_observed_at":"2026-08-22T11:41:06.722714Z","submitted_at":"2025-07-29T10:10:47Z","title":"DGP: A Dual-Granularity Prompting Framework for Fraud Detection with Graph-Enhanced LLMs","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2507.21653","snapshot_observed_at":"2026-08-02T07:06:20.220512Z","title":"DGP: A dual-granularity prompting framework for fraud detection with graph-enhanced LLMs, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:20.220512Z"},"links":{"cited_paper":"/paper/2507.21653","citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:85106bc1c95b54702a671b097fe45d3c319cdb73c18c550072a67d689b75bd5d","observation_id":"ebb48a25-053c-4371-b8c9-fc0236b50e85","resolution":{"observed_at":"2026-08-02T07:06:20.220512Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2502.18695","last_updated":"2025-02-25T23:15:16Z","snapshot_observed_at":"2026-08-21T05:28:54.772561Z","submitted_at":"2025-02-25T23:15:16Z","title":"Policy-as-Prompt: Rethinking Content Moderation in the Age of Large Language Models","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2502.18695","snapshot_observed_at":"2026-08-02T07:06:20.259839Z","title":"Taber, Andreas Damianou, and Mounia Lalmas","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:20.259839Z"},"links":{"cited_paper":"/paper/2502.18695","citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:c2fb3ae29b84bacff801cbe88da08df0a1cf1e15999fb49b034333649a636926","observation_id":"6e259eb8-1d3d-474c-878a-ab2eb6c5add3","resolution":{"observed_at":"2026-08-02T07:06:20.259839Z","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":"2026-08-02T07:06:20.314032Z","title":"Acomprehensivereviewof LLM-based content moderation: Advancements, challenges, and future directions, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:20.314032Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:4a64903b51c7ebb5c4f4957ceb5774608b861bc4a01bcce8b3f0dc18eefbe40b","observation_id":"4f428ee8-e385-423a-82d5-0fe31cf31f66","resolution":{"observed_at":"2026-08-02T07:06:20.314032Z","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":"2026-08-02T07:06:20.372316Z","title":"Security of LLM-based agents regarding attacks, defenses, and applications: A comprehensive survey, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:20.372316Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:0a009ae838e8b09f2e2fa53cefa2ddf8b7e83bdc6ee6d45ab942cfaaec3ae08d","observation_id":"15911858-b7cc-49d5-b079-a8313f68dcf2","resolution":{"observed_at":"2026-08-02T07:06:20.372316Z","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":"2026-08-02T07:06:20.404988Z","title":null,"venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:20.404988Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:ecd9a3be0ad7edf7464ac2d8485b3e4b49e44ef287147a7e009ecb5f65d7e6d1","observation_id":"7c73b2d6-e0f9-4750-a524-991b4090d151","resolution":{"observed_at":"2026-08-02T07:06:20.404988Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2409.03219","last_updated":"2025-06-01T05:04:14Z","snapshot_observed_at":"2026-08-17T06:16:58.507411Z","submitted_at":"2024-09-05T03:33:54Z","title":"Content Moderation by LLM: From Accuracy to Legitimacy","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2409.03219","snapshot_observed_at":"2026-08-02T07:06:20.486642Z","title":"Content moderation by LLM: From accuracy to legitimacy, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:20.486642Z"},"links":{"cited_paper":"/paper/2409.03219","citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:8c99b0e4995e538edfd584c6544bea8c3571fc80251087e13c94a7bf07f72c82","observation_id":"f5f6afd8-5445-4f59-b359-e842317cb521","resolution":{"observed_at":"2026-08-02T07:06:20.486642Z","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":"2026-08-02T07:06:20.609474Z","title":"TELUSdigitaltrustandsafetytrends2025.https://www.telusdigital","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:20.609474Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:01220cedc7d83de52766ffc666c9128d01f5969def55519992279858b3a291ee","observation_id":"ce04c4a3-6956-47e6-b5a2-8d341e12686f","resolution":{"observed_at":"2026-08-02T07:06:20.609474Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2504.00125","last_updated":"2025-03-31T18:19:41Z","snapshot_observed_at":"2026-08-18T15:37:24.853813Z","submitted_at":"2025-03-31T18:19:41Z","title":"LLMs for Explainable AI: A Comprehensive Survey","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2504.00125","snapshot_observed_at":"2026-08-02T07:06:20.676202Z","title":"LLMs for explainable AI: A comprehensive survey, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:20.676202Z"},"links":{"cited_paper":"/paper/2504.00125","citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:12f60aaf54b6d224df2ac7ffef0ec8830b7bec53009f147c0fb2a3126c47a48e","observation_id":"65481bcf-9fc8-41cf-969c-eabe7213b6f8","resolution":{"observed_at":"2026-08-02T07:06:20.676202Z","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":"2026-08-02T07:06:20.759370Z","title":"Dziemian, M","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:20.759370Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:694adbc7dc5a5cf63c6a2c53f0b8d96978f63bcbfa5567457cba159b4f5a0051","observation_id":"cced1deb-8dd4-491f-a597-f43af6c6d870","resolution":{"observed_at":"2026-08-02T07:06:20.759370Z","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":"2026-08-02T07:06:20.828834Z","title":"FRAUDLLM: Zero-shot fraud detection with large language models, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:20.828834Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:2dbea4cfac0c49f98d42c21b69c921896987a64f55c3aa5a88e0849d6dd40021","observation_id":"2cd0834a-5a30-4754-a148-2575fad8eb37","resolution":{"observed_at":"2026-08-02T07:06:20.828834Z","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":"2026-08-02T07:06:20.932411Z","title":"Reinforcement learning of large language models for interpretable credit card fraud detection, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:20.932411Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:ba61b1e7f985f9e9243ce312af320ee8dfdec5d1ec5c2053f4eb1792dc5d5cb8","observation_id":"0285e250-e781-41ab-8a06-fadab6ed034c","resolution":{"observed_at":"2026-08-02T07:06:20.932411Z","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":"2026-08-02T07:06:21.031188Z","title":"Telecom fraud detection based on large language models: A multi-role, multi-layer prompting strategy, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:21.031188Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:3092539dfe0918255361401241f0fccac9ab9ac55f21f1ab6ca8395f80cea724","observation_id":"719b3d5d-ea38-4fd2-8615-a9561b64e344","resolution":{"observed_at":"2026-08-02T07:06:21.031188Z","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":"2026-08-02T07:06:21.117359Z","title":"Telecom fraud recognition based on large language model neuron selection, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:21.117359Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:4410c4e31629bda42b53da4585e6212a82702b3e33cb3134be98e69cc0a4e06a","observation_id":"f6a45dc9-c2f5-474a-a61d-bdd97df6e08e","resolution":{"observed_at":"2026-08-02T07:06:21.117359Z","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":"2026-08-02T07:06:21.200183Z","title":"Can LLMs find fraudsters? multi-level LLM enhanced graph fraud detection, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:21.200183Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:e1e1bc29a3980aefdcfbf05991501e9770a9bc149f19a4421053c7bc72aa56b6","observation_id":"e8f27073-efe6-4fe8-a813-ae8bdf5489f7","resolution":{"observed_at":"2026-08-02T07:06:21.200183Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2507.14785","last_updated":"2025-10-29T15:56:28Z","snapshot_observed_at":"2026-08-13T13:02:53.578076Z","submitted_at":"2025-07-20T02:00:21Z","title":"Exploring the In-Context Learning Capabilities of LLMs for Money Laundering Detection in Financial Graphs","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2507.14785","snapshot_observed_at":"2026-08-02T07:06:21.268179Z","title":"Exploring the in-context learning capabilities of LLMs for money laundering detection in financial graphs, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:21.268179Z"},"links":{"cited_paper":"/paper/2507.14785","citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:93bb46935bff8db8c653d8b54d4c4ccce2da859f04732cfa437cd67fecaa7559","observation_id":"cd318301-4d26-4762-8215-0ac23ae52cc7","resolution":{"observed_at":"2026-08-02T07:06:21.268179Z","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":"2026-08-02T07:06:21.337198Z","title":null,"venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:21.337198Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:c7c4316f66e403014283035918c74a96dab26bfdf6b2822c9efce68e5bc4fd31","observation_id":"f1b77bc9-cce5-4688-9658-c58c32788714","resolution":{"observed_at":"2026-08-02T07:06:21.337198Z","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":"2026-08-02T07:06:21.419855Z","title":"Large language models reproduce racial stereotypes when used for text annotation, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:21.419855Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:0a5e2ca215ad6de56e88f7f8f636f7f8984d5f2bf61e380ed8ae8f72949eff6c","observation_id":"c4184226-9f46-42e2-a361-0d1698e8ea6a","resolution":{"observed_at":"2026-08-02T07:06:21.419855Z","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":"2026-08-02T07:06:21.512585Z","title":"Dialect vs demographics: Quantifying LLM bias from im- plicit linguistic signals vs","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":44,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:21.512585Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:e0635dcdfc49bae9e6ef7f14ab96b14c114c2891f72962e8d0446d81ce376d44","observation_id":"4ddeeb3c-5413-4247-8315-ab354a459f29","resolution":{"observed_at":"2026-08-02T07:06:21.512585Z","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":"2026-08-02T07:06:21.582965Z","title":"When to invoke: Refining LLM fairness with toxicity assessment, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":45,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:21.582965Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:23da94b2773cf35350739f46432a50efeb3ceeb0f569914797928a3bd5ee700c","observation_id":"6fc977fb-7f67-41b8-8df6-c99baac87642","resolution":{"observed_at":"2026-08-02T07:06:21.582965Z","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":"2026-08-02T07:06:21.647721Z","title":"Longitudinalmonitoring of LLM content moderation of social issues, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":46,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:21.647721Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:a9026b2783fefbcb8d7d1786bea96aa6c70946c9bf402b38c5fcce6ee0974758","observation_id":"913dd807-425e-44dd-99d1-4ef6f137f119","resolution":{"observed_at":"2026-08-02T07:06:21.647721Z","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":"2026-08-02T07:06:21.762707Z","title":"Prompt injection at- tacks in large language models and AI agent systems: A comprehensive review of vul- nerabilities, attack vectors, and defense mechanisms, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":47,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:21.762707Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:d65a5b7a5d3d000347b4341b532fa879ba667cb0c5990cd028478bc65a4e5edc","observation_id":"6c5899e9-3d0e-49b5-8164-cab5570410f0","resolution":{"observed_at":"2026-08-02T07:06:21.762707Z","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":"2026-08-02T07:06:21.824395Z","title":"Semantic chameleon: Corpus-dependent poisoning attacks and defenses in RAG systems, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":48,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:21.824395Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:ce984ec4f4f02a41206015a57b808ae641271a89f3d201356114a469dd270e61","observation_id":"a89ced59-3bf8-41dc-91d3-39ecb95b73ae","resolution":{"observed_at":"2026-08-02T07:06:21.824395Z","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":"2026-08-02T07:06:21.906580Z","title":"DataexfiltrationfromSlackAIviaindirectpromptinjection","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":49,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:21.906580Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:8dc7163df41bd862a8825d303efcaafc41e74c77c003d8ed17d7a8f6252ddd32","observation_id":"973fae18-0444-4723-bcae-447da30b5ff7","resolution":{"observed_at":"2026-08-02T07:06:21.906580Z","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":"2026-08-02T07:06:21.991812Z","title":"ServiceNow Now Assist AI agent vulnerability (BodySnatcher, cve-2025-12420),","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":50,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:21.991812Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:46787c0f279a6fcc571840c45845c9be71754fbe3d9b510cc4f5838d4b701178","observation_id":"5d332f9d-3f96-43e1-9061-42fa42a8bbc2","resolution":{"observed_at":"2026-08-02T07:06:21.991812Z","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":"2026-08-02T07:06:22.137016Z","title":"Detecting and analyzing prompt abuse in AI tools","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":51,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:22.137016Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:2ceed85d8da2ea24e152f10b4f70018c8cb466fe5c7aeedd4109ecfc23e4229b","observation_id":"bbc2ec67-3aee-4b1c-b2b1-6eeb1ad41084","resolution":{"observed_at":"2026-08-02T07:06:22.137016Z","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":"2026-08-02T07:06:22.211817Z","title":"OWASP GenAI exploit round-up report q1 2026.https: //genai.owasp.org/2026/04/14/owasp-genai-exploit-round-up-report-q1-2026/,","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":52,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:22.211817Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:d27cd1eeefd8207a97574d4844d97c5ca93b557d333000565bda3c5f57d3b4c5","observation_id":"94e8c56e-03f7-497b-a333-1c45800c7ea6","resolution":{"observed_at":"2026-08-02T07:06:22.211817Z","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":"2026-08-02T07:06:22.469183Z","title":null,"venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":53,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:22.469183Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:d749b72c6ba755648c940b0a40e69d56f665dd8e9c051164f588b3712df4546f","observation_id":"26974836-ea47-4e5a-b170-dafca61f7799","resolution":{"observed_at":"2026-08-02T07:06:22.469183Z","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":"2026-08-02T07:06:22.672309Z","title":"Breaking the chain: A causal analysis of LLM faithfulness to intermediate struc- tures, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":54,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:22.672309Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:0c03aa39758000891aa367184cc3997edaa9697f77c47c06e336c06e126c2c58","observation_id":"d9510f26-2698-4c69-a7ef-a294baf6610f","resolution":{"observed_at":"2026-08-02T07:06:22.672309Z","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":"2026-08-02T07:06:22.811347Z","title":"Multi-layered framework for LLM hallucination mitigation in high-stakes applications: A tutorial, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":55,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:22.811347Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:65ff79966c2dad3a7f7eecf212323fe849a4f0e92805e7991bdc195dea5aeb4e","observation_id":"af2e034e-aa42-4961-bb91-0115015c0a4b","resolution":{"observed_at":"2026-08-02T07:06:22.811347Z","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":"2026-08-02T07:06:22.317656Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":56,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:22.317656Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:485debc20c7561a9f049dc50fce33ee094e44c69312b6db1f23627b6d67f5476","observation_id":"fbf5cdef-a274-4ffd-8414-7bcbdfa6acaf","resolution":{"observed_at":"2026-08-02T07:06:22.317656Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2508.07872","last_updated":"2026-07-07T05:38:13Z","snapshot_observed_at":"2026-08-05T21:50:00.340490Z","submitted_at":"2025-08-11T11:43:34Z","title":"Unequal Uncertainty: Rethinking Algorithmic Interventions for Mitigating Discrimination from AI","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2508.07872","snapshot_observed_at":"2026-08-02T07:06:23.116966Z","title":"Un- equal uncertainty: Rethinking algorithmic interventions for mitigating discrimination from AI, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":57,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:23.116966Z"},"links":{"cited_paper":"/paper/2508.07872","citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:a5948ee5b368b7c3cfdd0d4eb60df45c30ea43c9147adc853efa4ecb6b8cdcec","observation_id":"a089d754-05d9-4b06-be52-081688b0e346","resolution":{"observed_at":"2026-08-02T07:06:23.116966Z","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":"2026-08-02T07:06:23.187092Z","title":"Can large language models automate phishing warning explanations? a controlled experiment on effectiveness and user perception, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":58,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:23.187092Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:515d76507d5e38b594ca2c9dea773924cc78d3e9752438bc2ba357ef6b1bb74a","observation_id":"0c594956-b7ad-415e-bb25-583fb74fd094","resolution":{"observed_at":"2026-08-02T07:06:23.187092Z","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":"2026-08-02T07:06:23.268238Z","title":null,"venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":59,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:23.268238Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:d78c4c0a5919b1785a5c1f9a875be5a55e82bce0804e376377604983476bf899","observation_id":"71ec504e-d121-4036-a127-f028d0264c18","resolution":{"observed_at":"2026-08-02T07:06:23.268238Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2502.06884","last_updated":"2025-02-08T21:30:41Z","snapshot_observed_at":"2026-08-19T23:48:15.873299Z","submitted_at":"2025-02-08T21:30:41Z","title":"Learning Conformal Abstention Policies for Adaptive Risk Management in Large Language and Vision-Language Models","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2502.06884","snapshot_observed_at":"2026-08-02T07:06:22.993466Z","title":"Learning conformal abstention policies for adaptive risk manage- ment in large language and vision-language models, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":60,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:22.993466Z"},"links":{"cited_paper":"/paper/2502.06884","citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:f81a2eae9761f4abe9c8c1bc4142b15f97fa0e4945115acf76e94f73d68a442e","observation_id":"d42a87e4-4fdd-48a4-b1dc-927df1384132","resolution":{"observed_at":"2026-08-02T07:06:22.993466Z","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":"2026-08-02T07:06:23.361289Z","title":null,"venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":61,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:23.361289Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:93d4172b7cd634b729d5c180792c2616bf30d57e553c1ac6d381628872b8ac8b","observation_id":"06987d35-ad35-477b-abef-5fe29278b232","resolution":{"observed_at":"2026-08-02T07:06:23.361289Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2506.21443","last_updated":"2025-06-26T16:29:45Z","snapshot_observed_at":"2026-08-18T00:11:37.469583Z","submitted_at":"2025-06-26T16:29:45Z","title":"Domain Knowledge-Enhanced LLMs for Fraud and Concept Drift Detection","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2506.21443","snapshot_observed_at":"2026-08-02T07:06:23.445924Z","title":"Domain knowledge-enhanced LLMs for fraud and concept drift detection, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":62,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:23.445924Z"},"links":{"cited_paper":"/paper/2506.21443","citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:8b723e2d9a3521d36dadc76ba3091de3d139881822ef234c3e0a5af624b662e9","observation_id":"48b12ee2-ffc5-4171-bfc3-37178c625609","resolution":{"observed_at":"2026-08-02T07:06:23.445924Z","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":"2026-08-02T07:06:23.512024Z","title":"LLM performance predictors: Learning when to escalate in hybrid human-AI moderation systems, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":63,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:23.512024Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:ea716998acbb764ab21e8eaf6248bef48b57752aecfc0717d83cfce52ade4595","observation_id":"afe47d99-b6ea-4d9c-9bf1-4de9fbdd4bd6","resolution":{"observed_at":"2026-08-02T07:06:23.512024Z","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":"2026-08-02T07:06:23.322568Z","title":null,"venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":64,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:23.322568Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:41e303fd026ed769eca96475bbf9282075cd09571f5b21e85ddbf79d70c30600","observation_id":"e19b6f40-5b9c-4ded-bbc0-0269db94434f","resolution":{"observed_at":"2026-08-02T07:06:23.322568Z","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":"2026-08-02T07:06:23.669203Z","title":"UCCI: Calibrated uncertainty for cost-optimal LLM cascade routing, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":65,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:23.669203Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:a597ecacf299179f0675d0e58bc26f185cbca7d9a40aaa46626127059596796d","observation_id":"01608456-5d0a-46ac-b486-83e630f2588f","resolution":{"observed_at":"2026-08-02T07:06:23.669203Z","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":"2026-08-02T07:06:23.755495Z","title":"Compliance-scored best-of-N guardrail or- chestration for multimodal document generation in payments dispute defense, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":66,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:23.755495Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:cc171b75a39b6f93d158169748f3722bd4e9398d7ee32b1c95b4d58c477b6798","observation_id":"2931a3bf-d6d4-4ec2-9941-87eb04599ab3","resolution":{"observed_at":"2026-08-02T07:06:23.755495Z","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":"2026-08-02T07:06:23.813683Z","title":"Robust and efficient guardrails with latent reasoning, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":67,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:23.813683Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:e341ccdf84481709c39b7051b036c6a7d9379f65755a4e09eac65885ea69431f","observation_id":"e2f4c2c1-12f1-446a-8119-9ddc50266a1e","resolution":{"observed_at":"2026-08-02T07:06:23.813683Z","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":"2026-08-02T07:06:23.598570Z","title":"Redefining AI red teaming in the agentic era: From weeks to hours, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":68,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:23.598570Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:c02ef86aa04bf3271034b50a6d7cb780dba6c27511450fb1c56eba910b7f2039","observation_id":"1a68ae49-12bd-49af-bc0e-3d70b3d995ad","resolution":{"observed_at":"2026-08-02T07:06:23.598570Z","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":"2026-08-02T07:06:23.993917Z","title":"AI agents may always fall for prompt injections,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":69,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:23.993917Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:6c9fc6a5e01dd45c147e0685fda14124e8dcbfaceabcba8cb3c5bcec55d08175","observation_id":"904009e3-d20f-4597-b98b-e3b9edf43940","resolution":{"observed_at":"2026-08-02T07:06:23.993917Z","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":"2026-08-02T07:06:23.908774Z","title":"SAGE: An LLM- driven self reflective agentic framework for fraud detection, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":72,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:23.908774Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:bebc8d5a252edc97e3cd432c26eae8d5256c22b871121088c8ed8a988d279380","observation_id":"e82dff18-7ea6-4f72-b25f-aa666543e3d5","resolution":{"observed_at":"2026-08-02T07:06:23.908774Z","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":"2026-08-02T07:06:24.001849Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":74,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:24.001849Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:13a72de15baa9bb3a91710216ebbe0e9d8cd382649caa566f7f7528eecd85884","observation_id":"d14cceb2-66e2-4b8b-bea1-4f79f750fde6","resolution":{"observed_at":"2026-08-02T07:06:24.001849Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2312.02003","last_updated":"2024-03-20T19:00:24Z","snapshot_observed_at":"2026-08-19T12:53:10.765507Z","submitted_at":"2023-12-04T16:25:18Z","title":"A Survey on Large Language Model (LLM) Security and Privacy: The Good, the Bad, and the Ugly","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2312.02003","snapshot_observed_at":"2026-08-02T07:06:19.581859Z","title":null,"venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":2024,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:19.581859Z"},"links":{"cited_paper":"/paper/2312.02003","citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:e32b423240c3c86ba5bd0451dc5ac7bd565ed240e516d2e38492a097df1d58ea","observation_id":"75363083-5c36-481d-903b-e2745f9bb9fd","resolution":{"observed_at":"2026-08-02T07:06:19.581859Z","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":"2026-08-02T07:06:22.059647Z","title":null,"venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":2025,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:22.059647Z"},"links":{"citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:225c4b9e7b04854f8b7e2e492db7d13e3642200df5f2eee021784613f956cb94","observation_id":"3aecad44-4622-4970-958c-c2e30e62020c","resolution":{"observed_at":"2026-08-02T07:06:22.059647Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2601.19684","last_updated":"2026-04-28T08:41:50Z","snapshot_observed_at":"2026-08-08T13:12:00.177861Z","submitted_at":"2026-01-27T15:01:37Z","title":"LLM-Assisted Authentication and Fraud Detection","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2601.19684","snapshot_observed_at":"2026-08-02T07:06:18.241831Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows","version":1},"reference_index":2026,"source":"pdf_text","source_observed_at":"2026-08-02T07:06:18.241831Z"},"links":{"cited_paper":"/paper/2601.19684","citing_paper":"/paper/2607.13078"},"observation_digest":"sha256:74f6f23359a747b9226bdac0e0d48e58f411c4e1c45dc6132a57882cfe2bb94b","observation_id":"7c1ae580-f064-4936-a0fd-593d47fec544","resolution":{"observed_at":"2026-08-02T07:06:18.241831Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"paper":{"arxiv_id":"2607.13078","last_updated":"2026-07-12T22:34:02Z","latest_version":1,"primary_category":"cs.CR","snapshot_observed_at":"2026-08-09T02:07:52.896888Z","submitted_at":"2026-07-12T22:34:02Z","title":"Operational Evidence Gaps for LLMs in Fraud Detection and Trust-and-Safety Workflows"},"reference_resolution":{"displayed":74,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":73,"verified_exact":1,"verified_fuzzy":0},"total_outbound_references":74},"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":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+00:00","source":"retraction_watch"}],"thesis":"As of 24 August 2026, this Paper Citation Record lists 74 of 74 outbound references and 0 inbound Pith citation observations for arXiv:2607.13078."}