{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:45C3OQ32U6Q5XDEWZO4GKM3AXZ","short_pith_number":"pith:45C3OQ32","schema_version":"1.0","canonical_sha256":"e745b7437aa7a1db8c96cbb8653360be45e3a9d57395fceac1eb82902d85898d","source":{"kind":"arxiv","id":"2603.11331","version":3},"attestation_state":"computed","paper":{"title":"Jailbreak Scaling Laws for Large Language Models: Polynomial-Exponential Crossover","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Strong adversarial prompt injections turn slow polynomial growth of jailbreak success into exponential growth with more inference samples.","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Annesya Banerjee, Cengiz Pehlevan, Indranil Halder","submitted_at":"2026-03-11T21:48:03Z","abstract_excerpt":"Adversarial attacks can reliably steer safety-aligned large language models toward unsafe behavior. Empirically, we find that adversarial prompt-injection attacks can amplify attack success rate from the slow polynomial growth observed without injection to exponential growth with the number of inference-time samples. We first identify a minimal statistical mechanism for these two regimes by giving a small set of assumptions on the distribution of safe generation across contexts under which both scaling laws follow. To explain this phenomenon further, we propose a theoretical generative model o"},"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":true},"canonical_record":{"source":{"id":"2603.11331","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-03-11T21:48:03Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b19d9a16ee1e13b496c430be41e2c32586c75a70fac747d5c292d330614277e5","abstract_canon_sha256":"c8b0f935c930efed2220208d4f5886b601e689720c7b0ebbb4674ba75354df7b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T02:05:42.784457Z","signature_b64":"a5wGjQP+HNJ83WoqnOfJUgC8B/mO9JJ6ISZwADC+BMrB265obXYBwevKEIlViJWzht10yKceHnyDpoxO/isjCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e745b7437aa7a1db8c96cbb8653360be45e3a9d57395fceac1eb82902d85898d","last_reissued_at":"2026-05-29T02:05:42.783753Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T02:05:42.783753Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Jailbreak Scaling Laws for Large Language Models: Polynomial-Exponential Crossover","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Strong adversarial prompt injections turn slow polynomial growth of jailbreak success into exponential growth with more inference samples.","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Annesya Banerjee, Cengiz Pehlevan, Indranil Halder","submitted_at":"2026-03-11T21:48:03Z","abstract_excerpt":"Adversarial attacks can reliably steer safety-aligned large language models toward unsafe behavior. Empirically, we find that adversarial prompt-injection attacks can amplify attack success rate from the slow polynomial growth observed without injection to exponential growth with the number of inference-time samples. We first identify a minimal statistical mechanism for these two regimes by giving a small set of assumptions on the distribution of safe generation across contexts under which both scaling laws follow. To explain this phenomenon further, we propose a theoretical generative model o"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"strong adversarial prompt-injection attacks can amplify attack success rate from the slow polynomial growth observed without injection to exponential growth with the number of inference-time samples","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"a small set of assumptions on the distribution of safe generation across contexts under which both scaling laws follow","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Strong adversarial prompt injections shift jailbreak success scaling from polynomial to exponential with more inference samples, derived from a spin-glass generative model of LLM outputs.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Strong adversarial prompt injections turn slow polynomial growth of jailbreak success into exponential growth with more inference samples.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"62c11c0d1a70bcc0f3c9f75dcc2ee35722dead7734d613a721ab458319785aca"},"source":{"id":"2603.11331","kind":"arxiv","version":3},"verdict":{"id":"46c5e45c-d2dd-4f7c-baf8-c678a02e2eb5","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T12:37:31.020882Z","strongest_claim":"strong adversarial prompt-injection attacks can amplify attack success rate from the slow polynomial growth observed without injection to exponential growth with the number of inference-time samples","one_line_summary":"Strong adversarial prompt injections shift jailbreak success scaling from polynomial to exponential with more inference samples, derived from a spin-glass generative model of LLM outputs.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"a small set of assumptions on the distribution of safe generation across contexts under which both scaling laws follow","pith_extraction_headline":"Strong adversarial prompt injections turn slow polynomial growth of jailbreak success into exponential growth with more inference samples."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2603.11331/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":2,"snapshot_sha256":"1fe285041d1e5885d3400c8cde56fc8851d3c4936f4553eea7c537da583f13f4"},"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":"2603.11331","created_at":"2026-05-29T02:05:42.783830+00:00"},{"alias_kind":"arxiv_version","alias_value":"2603.11331v3","created_at":"2026-05-29T02:05:42.783830+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.11331","created_at":"2026-05-29T02:05:42.783830+00:00"},{"alias_kind":"pith_short_12","alias_value":"45C3OQ32U6Q5","created_at":"2026-05-29T02:05:42.783830+00:00"},{"alias_kind":"pith_short_16","alias_value":"45C3OQ32U6Q5XDEW","created_at":"2026-05-29T02:05:42.783830+00:00"},{"alias_kind":"pith_short_8","alias_value":"45C3OQ32","created_at":"2026-05-29T02:05:42.783830+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/45C3OQ32U6Q5XDEWZO4GKM3AXZ","json":"https://pith.science/pith/45C3OQ32U6Q5XDEWZO4GKM3AXZ.json","graph_json":"https://pith.science/api/pith-number/45C3OQ32U6Q5XDEWZO4GKM3AXZ/graph.json","events_json":"https://pith.science/api/pith-number/45C3OQ32U6Q5XDEWZO4GKM3AXZ/events.json","paper":"https://pith.science/paper/45C3OQ32"},"agent_actions":{"view_html":"https://pith.science/pith/45C3OQ32U6Q5XDEWZO4GKM3AXZ","download_json":"https://pith.science/pith/45C3OQ32U6Q5XDEWZO4GKM3AXZ.json","view_paper":"https://pith.science/paper/45C3OQ32","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2603.11331&json=true","fetch_graph":"https://pith.science/api/pith-number/45C3OQ32U6Q5XDEWZO4GKM3AXZ/graph.json","fetch_events":"https://pith.science/api/pith-number/45C3OQ32U6Q5XDEWZO4GKM3AXZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/45C3OQ32U6Q5XDEWZO4GKM3AXZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/45C3OQ32U6Q5XDEWZO4GKM3AXZ/action/storage_attestation","attest_author":"https://pith.science/pith/45C3OQ32U6Q5XDEWZO4GKM3AXZ/action/author_attestation","sign_citation":"https://pith.science/pith/45C3OQ32U6Q5XDEWZO4GKM3AXZ/action/citation_signature","submit_replication":"https://pith.science/pith/45C3OQ32U6Q5XDEWZO4GKM3AXZ/action/replication_record"}},"created_at":"2026-05-29T02:05:42.783830+00:00","updated_at":"2026-05-29T02:05:42.783830+00:00"}