{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:BEKQ5EY6FVHMABSI2BQIMKUIFM","short_pith_number":"pith:BEKQ5EY6","schema_version":"1.0","canonical_sha256":"09150e931e2d4ec00648d060862a882b0602fca6ac480f0ce661aece6467efb5","source":{"kind":"arxiv","id":"2606.25476","version":1},"attestation_state":"computed","paper":{"title":"A Red Teaming Framework for Large Language Models: A Case Study on Faithfulness Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Abrar Alotaibi, Moataz Ahmed, Raed Mughus","submitted_at":"2026-06-24T07:00:53Z","abstract_excerpt":"Large language models (LLMs) have demonstrated remarkable performance across natural language processing tasks, yet their deployment in high-stakes applications raises critical concerns regarding reliability, safety, and trustworthiness. In this paper, we present a red teaming framework that systematically uncovers vulnerabilities in LLM outputs. Our approach employs a novel multi-role architecture comprising target, attacker, and jury models. The attackers generate increasingly effective adversarial prompts while the jury rigorously evaluates response accuracy and consistency across tasks. In"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2606.25476","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-24T07:00:53Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9b71c18ce21bb91cf0d20bd38cf5787840c532f6a57ff494c8bf3d4c69217084","abstract_canon_sha256":"cac3cccae72c1bd233cf8ee5239195385b274c2fd0e6ae6383d2014e9903d63d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-25T01:18:06.382709Z","signature_b64":"80nEfNU6ynw8Ft8YotoJAx+UFD/M8wTjiJKl5h1WPw/0WVQl8oYlK1COnNDwAM67Cod4jov8Ksbk1HcfVQD/Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"09150e931e2d4ec00648d060862a882b0602fca6ac480f0ce661aece6467efb5","last_reissued_at":"2026-06-25T01:18:06.382376Z","signature_status":"signed_v1","first_computed_at":"2026-06-25T01:18:06.382376Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Red Teaming Framework for Large Language Models: A Case Study on Faithfulness Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Abrar Alotaibi, Moataz Ahmed, Raed Mughus","submitted_at":"2026-06-24T07:00:53Z","abstract_excerpt":"Large language models (LLMs) have demonstrated remarkable performance across natural language processing tasks, yet their deployment in high-stakes applications raises critical concerns regarding reliability, safety, and trustworthiness. In this paper, we present a red teaming framework that systematically uncovers vulnerabilities in LLM outputs. Our approach employs a novel multi-role architecture comprising target, attacker, and jury models. The attackers generate increasingly effective adversarial prompts while the jury rigorously evaluates response accuracy and consistency across tasks. In"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.25476","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2606.25476/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2606.25476","created_at":"2026-06-25T01:18:06.382429+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.25476v1","created_at":"2026-06-25T01:18:06.382429+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.25476","created_at":"2026-06-25T01:18:06.382429+00:00"},{"alias_kind":"pith_short_12","alias_value":"BEKQ5EY6FVHM","created_at":"2026-06-25T01:18:06.382429+00:00"},{"alias_kind":"pith_short_16","alias_value":"BEKQ5EY6FVHMABSI","created_at":"2026-06-25T01:18:06.382429+00:00"},{"alias_kind":"pith_short_8","alias_value":"BEKQ5EY6","created_at":"2026-06-25T01:18:06.382429+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BEKQ5EY6FVHMABSI2BQIMKUIFM","json":"https://pith.science/pith/BEKQ5EY6FVHMABSI2BQIMKUIFM.json","graph_json":"https://pith.science/api/pith-number/BEKQ5EY6FVHMABSI2BQIMKUIFM/graph.json","events_json":"https://pith.science/api/pith-number/BEKQ5EY6FVHMABSI2BQIMKUIFM/events.json","paper":"https://pith.science/paper/BEKQ5EY6"},"agent_actions":{"view_html":"https://pith.science/pith/BEKQ5EY6FVHMABSI2BQIMKUIFM","download_json":"https://pith.science/pith/BEKQ5EY6FVHMABSI2BQIMKUIFM.json","view_paper":"https://pith.science/paper/BEKQ5EY6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.25476&json=true","fetch_graph":"https://pith.science/api/pith-number/BEKQ5EY6FVHMABSI2BQIMKUIFM/graph.json","fetch_events":"https://pith.science/api/pith-number/BEKQ5EY6FVHMABSI2BQIMKUIFM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BEKQ5EY6FVHMABSI2BQIMKUIFM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BEKQ5EY6FVHMABSI2BQIMKUIFM/action/storage_attestation","attest_author":"https://pith.science/pith/BEKQ5EY6FVHMABSI2BQIMKUIFM/action/author_attestation","sign_citation":"https://pith.science/pith/BEKQ5EY6FVHMABSI2BQIMKUIFM/action/citation_signature","submit_replication":"https://pith.science/pith/BEKQ5EY6FVHMABSI2BQIMKUIFM/action/replication_record"}},"created_at":"2026-06-25T01:18:06.382429+00:00","updated_at":"2026-06-25T01:18:06.382429+00:00"}