{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:KLAEZOL7FFBRTXRGLSC3LZMK72","short_pith_number":"pith:KLAEZOL7","schema_version":"1.0","canonical_sha256":"52c04cb97f294319de265c85b5e58afea2ffa321e60dfb297ef8a26ee3bdb2f3","source":{"kind":"arxiv","id":"2501.02997","version":1},"attestation_state":"computed","paper":{"title":"CALM: Curiosity-Driven Auditing for Large Language Models","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Chao Shen, Cong Wang, Longxiang Wang, Xiang Zheng, Xingjun Ma, Yi Liu","submitted_at":"2025-01-06T13:14:34Z","abstract_excerpt":"Auditing Large Language Models (LLMs) is a crucial and challenging task. In this study, we focus on auditing black-box LLMs without access to their parameters, only to the provided service. We treat this type of auditing as a black-box optimization problem where the goal is to automatically uncover input-output pairs of the target LLMs that exhibit illegal, immoral, or unsafe behaviors. For instance, we may seek a non-toxic input that the target LLM responds to with a toxic output or an input that induces the hallucinative response from the target LLM containing politically sensitive individua"},"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":"2501.02997","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.AI","submitted_at":"2025-01-06T13:14:34Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"a20ca33f741e794c2431ea3719a66fd12240f8e7fd5456320332b1f2934eed4d","abstract_canon_sha256":"f8258880987259824177fc2dabdf5f4b9cf44cf18d45f49722fe6927ca5a009b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:57:28.936507Z","signature_b64":"Avff/6TccNUVWQio3OOWrhYSAEvXmIBERhMjnHXIm3gpItXiUZCmXz4s4wvBzEGp/oCyjwKsNOrp2Qiy3OKKDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"52c04cb97f294319de265c85b5e58afea2ffa321e60dfb297ef8a26ee3bdb2f3","last_reissued_at":"2026-07-05T09:57:28.936052Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:57:28.936052Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CALM: Curiosity-Driven Auditing for Large Language Models","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Chao Shen, Cong Wang, Longxiang Wang, Xiang Zheng, Xingjun Ma, Yi Liu","submitted_at":"2025-01-06T13:14:34Z","abstract_excerpt":"Auditing Large Language Models (LLMs) is a crucial and challenging task. In this study, we focus on auditing black-box LLMs without access to their parameters, only to the provided service. We treat this type of auditing as a black-box optimization problem where the goal is to automatically uncover input-output pairs of the target LLMs that exhibit illegal, immoral, or unsafe behaviors. For instance, we may seek a non-toxic input that the target LLM responds to with a toxic output or an input that induces the hallucinative response from the target LLM containing politically sensitive individua"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.02997","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/2501.02997/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":"2501.02997","created_at":"2026-07-05T09:57:28.936112+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.02997v1","created_at":"2026-07-05T09:57:28.936112+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.02997","created_at":"2026-07-05T09:57:28.936112+00:00"},{"alias_kind":"pith_short_12","alias_value":"KLAEZOL7FFBR","created_at":"2026-07-05T09:57:28.936112+00:00"},{"alias_kind":"pith_short_16","alias_value":"KLAEZOL7FFBRTXRG","created_at":"2026-07-05T09:57:28.936112+00:00"},{"alias_kind":"pith_short_8","alias_value":"KLAEZOL7","created_at":"2026-07-05T09:57:28.936112+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.04302","citing_title":"RedRFT: A Light-Weight Benchmark for Reinforcement Fine-Tuning-Based Red Teaming","ref_index":14,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KLAEZOL7FFBRTXRGLSC3LZMK72","json":"https://pith.science/pith/KLAEZOL7FFBRTXRGLSC3LZMK72.json","graph_json":"https://pith.science/api/pith-number/KLAEZOL7FFBRTXRGLSC3LZMK72/graph.json","events_json":"https://pith.science/api/pith-number/KLAEZOL7FFBRTXRGLSC3LZMK72/events.json","paper":"https://pith.science/paper/KLAEZOL7"},"agent_actions":{"view_html":"https://pith.science/pith/KLAEZOL7FFBRTXRGLSC3LZMK72","download_json":"https://pith.science/pith/KLAEZOL7FFBRTXRGLSC3LZMK72.json","view_paper":"https://pith.science/paper/KLAEZOL7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.02997&json=true","fetch_graph":"https://pith.science/api/pith-number/KLAEZOL7FFBRTXRGLSC3LZMK72/graph.json","fetch_events":"https://pith.science/api/pith-number/KLAEZOL7FFBRTXRGLSC3LZMK72/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KLAEZOL7FFBRTXRGLSC3LZMK72/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KLAEZOL7FFBRTXRGLSC3LZMK72/action/storage_attestation","attest_author":"https://pith.science/pith/KLAEZOL7FFBRTXRGLSC3LZMK72/action/author_attestation","sign_citation":"https://pith.science/pith/KLAEZOL7FFBRTXRGLSC3LZMK72/action/citation_signature","submit_replication":"https://pith.science/pith/KLAEZOL7FFBRTXRGLSC3LZMK72/action/replication_record"}},"created_at":"2026-07-05T09:57:28.936112+00:00","updated_at":"2026-07-05T09:57:28.936112+00:00"}