{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:FAFULRYCEYMKTHDGNJIZ2VQKNP","short_pith_number":"pith:FAFULRYC","schema_version":"1.0","canonical_sha256":"280b45c7022618a99c666a519d560a6bf9b6def987b0c0d102118cbfbedc0aa7","source":{"kind":"arxiv","id":"2505.24232","version":1},"attestation_state":"computed","paper":{"title":"From Hallucinations to Jailbreaks: Rethinking the Vulnerability of Large Foundation Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.CV","authors_text":"Haibo Jin, Haohan Wang, Man Luo, Peiran Wang, Peiyan Zhang","submitted_at":"2025-05-30T05:48:50Z","abstract_excerpt":"Large foundation models (LFMs) are susceptible to two distinct vulnerabilities: hallucinations and jailbreak attacks. While typically studied in isolation, we observe that defenses targeting one often affect the other, hinting at a deeper connection.\n  We propose a unified theoretical framework that models jailbreaks as token-level optimization and hallucinations as attention-level optimization. Within this framework, we establish two key propositions: (1) \\textit{Similar Loss Convergence} - the loss functions for both vulnerabilities converge similarly when optimizing for target-specific outp"},"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":"2505.24232","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-30T05:48:50Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"e814a84a7b35f35781da7d98dba13f5e132e2bbf010bc47fffacbe265e320ac2","abstract_canon_sha256":"981c8656f8f111f1db5c83d7814f91f2e1ebc22ee63d01dd28486f3f699ec19e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:12:44.727409Z","signature_b64":"WjW5legbzxRyC2dQTAYB8VraR0lh/pV5xhO0P+RK/k5Pmq0TaFL5UtwQ6egQpYtRNfmuD59rUx2HQo9OS9GjBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"280b45c7022618a99c666a519d560a6bf9b6def987b0c0d102118cbfbedc0aa7","last_reissued_at":"2026-07-05T11:12:44.726873Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:12:44.726873Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"From Hallucinations to Jailbreaks: Rethinking the Vulnerability of Large Foundation Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.CV","authors_text":"Haibo Jin, Haohan Wang, Man Luo, Peiran Wang, Peiyan Zhang","submitted_at":"2025-05-30T05:48:50Z","abstract_excerpt":"Large foundation models (LFMs) are susceptible to two distinct vulnerabilities: hallucinations and jailbreak attacks. While typically studied in isolation, we observe that defenses targeting one often affect the other, hinting at a deeper connection.\n  We propose a unified theoretical framework that models jailbreaks as token-level optimization and hallucinations as attention-level optimization. Within this framework, we establish two key propositions: (1) \\textit{Similar Loss Convergence} - the loss functions for both vulnerabilities converge similarly when optimizing for target-specific outp"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.24232","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/2505.24232/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":"2505.24232","created_at":"2026-07-05T11:12:44.726940+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.24232v1","created_at":"2026-07-05T11:12:44.726940+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.24232","created_at":"2026-07-05T11:12:44.726940+00:00"},{"alias_kind":"pith_short_12","alias_value":"FAFULRYCEYMK","created_at":"2026-07-05T11:12:44.726940+00:00"},{"alias_kind":"pith_short_16","alias_value":"FAFULRYCEYMKTHDG","created_at":"2026-07-05T11:12:44.726940+00:00"},{"alias_kind":"pith_short_8","alias_value":"FAFULRYC","created_at":"2026-07-05T11:12:44.726940+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/FAFULRYCEYMKTHDGNJIZ2VQKNP","json":"https://pith.science/pith/FAFULRYCEYMKTHDGNJIZ2VQKNP.json","graph_json":"https://pith.science/api/pith-number/FAFULRYCEYMKTHDGNJIZ2VQKNP/graph.json","events_json":"https://pith.science/api/pith-number/FAFULRYCEYMKTHDGNJIZ2VQKNP/events.json","paper":"https://pith.science/paper/FAFULRYC"},"agent_actions":{"view_html":"https://pith.science/pith/FAFULRYCEYMKTHDGNJIZ2VQKNP","download_json":"https://pith.science/pith/FAFULRYCEYMKTHDGNJIZ2VQKNP.json","view_paper":"https://pith.science/paper/FAFULRYC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.24232&json=true","fetch_graph":"https://pith.science/api/pith-number/FAFULRYCEYMKTHDGNJIZ2VQKNP/graph.json","fetch_events":"https://pith.science/api/pith-number/FAFULRYCEYMKTHDGNJIZ2VQKNP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FAFULRYCEYMKTHDGNJIZ2VQKNP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FAFULRYCEYMKTHDGNJIZ2VQKNP/action/storage_attestation","attest_author":"https://pith.science/pith/FAFULRYCEYMKTHDGNJIZ2VQKNP/action/author_attestation","sign_citation":"https://pith.science/pith/FAFULRYCEYMKTHDGNJIZ2VQKNP/action/citation_signature","submit_replication":"https://pith.science/pith/FAFULRYCEYMKTHDGNJIZ2VQKNP/action/replication_record"}},"created_at":"2026-07-05T11:12:44.726940+00:00","updated_at":"2026-07-05T11:12:44.726940+00:00"}