{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:355I2WDQVG6H4AGLMZJ7MKX76B","short_pith_number":"pith:355I2WDQ","schema_version":"1.0","canonical_sha256":"df7a8d5870a9bc7e00cb6653f62afff0569904ce0232ad2a2eebd5a5370e9839","source":{"kind":"arxiv","id":"2605.10764","version":2},"attestation_state":"computed","paper":{"title":"Break the Brake, Not the Wheel: Untargeted Jailbreak via Entropy Maximization","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"Maximizing entropy at high-entropy refusal tokens flips VLM outputs to harmful content without fixed targets and improves cross-model transfer.","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Jing Zhang, Jinhong Ni, Mengqi He, Shu Zou, Weikang Li, Xin Shen, Xinyu Tian, Xuesong Li, Zhaoyuan Yang","submitted_at":"2026-05-11T15:59:02Z","abstract_excerpt":"Recent studies show that gradient-based universal image jailbreaks on vision-language models (VLMs) exhibit little or no cross-model transferability, casting doubt on the feasibility of transferable multimodal jailbreaks. We revisit this conclusion under a strictly untargeted threat model without enforcing a fixed prefix or response pattern. Our preliminary experiment reveals that refusal behavior concentrates at high-entropy tokens during autoregressive decoding, and non-refusal tokens already carry substantial probability mass among the top-ranked candidates before attack. Motivated by this "},"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":"2605.10764","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-11T15:59:02Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f5c1e9f618941bc18ce7c568364bfaf7809f3be7982f71f7ae19ea60bea8461b","abstract_canon_sha256":"400567a152464ca902375c7e3e7482132176d8dab8640e218fc4a55334e5902c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-26T01:03:33.239558Z","signature_b64":"pnmeki/e9QsItZBGL7Y313FNiH89amJamnenCOu5p70IDEyUEYPGf+9cJvt6Jhjfs7O2KV+OnYXdP4Lqtjs5Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"df7a8d5870a9bc7e00cb6653f62afff0569904ce0232ad2a2eebd5a5370e9839","last_reissued_at":"2026-05-26T01:03:33.238765Z","signature_status":"signed_v1","first_computed_at":"2026-05-26T01:03:33.238765Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Break the Brake, Not the Wheel: Untargeted Jailbreak via Entropy Maximization","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"Maximizing entropy at high-entropy refusal tokens flips VLM outputs to harmful content without fixed targets and improves cross-model transfer.","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Jing Zhang, Jinhong Ni, Mengqi He, Shu Zou, Weikang Li, Xin Shen, Xinyu Tian, Xuesong Li, Zhaoyuan Yang","submitted_at":"2026-05-11T15:59:02Z","abstract_excerpt":"Recent studies show that gradient-based universal image jailbreaks on vision-language models (VLMs) exhibit little or no cross-model transferability, casting doubt on the feasibility of transferable multimodal jailbreaks. We revisit this conclusion under a strictly untargeted threat model without enforcing a fixed prefix or response pattern. Our preliminary experiment reveals that refusal behavior concentrates at high-entropy tokens during autoregressive decoding, and non-refusal tokens already carry substantial probability mass among the top-ranked candidates before attack. Motivated by this "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Our experimental results indicate that the limited transferability primarily stems from overly constrained optimization objectives.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"Refusal behavior concentrates at high-entropy tokens during autoregressive decoding, and non-refusal tokens already carry substantial probability mass among the top-ranked candidates before attack.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"UJEM-KL improves cross-model transferability of untargeted jailbreaks on vision-language models by maximizing entropy at decision tokens instead of forcing specific outputs.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Maximizing entropy at high-entropy refusal tokens flips VLM outputs to harmful content without fixed targets and improves cross-model transfer.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"fc5d00d99395f5bbb300cf1cfa64b688303d889a24af5dfbb3c1036bb80ba050"},"source":{"id":"2605.10764","kind":"arxiv","version":2},"verdict":{"id":"a3237278-d062-425b-986f-5d6978ec1dd0","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-12T03:32:55.001710Z","strongest_claim":"Our experimental results indicate that the limited transferability primarily stems from overly constrained optimization objectives.","one_line_summary":"UJEM-KL improves cross-model transferability of untargeted jailbreaks on vision-language models by maximizing entropy at decision tokens instead of forcing specific outputs.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"Refusal behavior concentrates at high-entropy tokens during autoregressive decoding, and non-refusal tokens already carry substantial probability mass among the top-ranked candidates before attack.","pith_extraction_headline":"Maximizing entropy at high-entropy refusal tokens flips VLM outputs to harmful content without fixed targets and improves cross-model transfer."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.10764/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"claim_evidence","ran_at":"2026-05-20T05:22:00.400417Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T14:36:17.535678Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T11:01:16.758894Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T09:00:03.992680Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"4eae519190e07d72913539afbad17157daf12591da49f26910a6c2bb73fd4daf"},"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":"2605.10764","created_at":"2026-05-26T01:03:33.238879+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.10764v2","created_at":"2026-05-26T01:03:33.238879+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.10764","created_at":"2026-05-26T01:03:33.238879+00:00"},{"alias_kind":"pith_short_12","alias_value":"355I2WDQVG6H","created_at":"2026-05-26T01:03:33.238879+00:00"},{"alias_kind":"pith_short_16","alias_value":"355I2WDQVG6H4AGL","created_at":"2026-05-26T01:03:33.238879+00:00"},{"alias_kind":"pith_short_8","alias_value":"355I2WDQ","created_at":"2026-05-26T01:03:33.238879+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/355I2WDQVG6H4AGLMZJ7MKX76B","json":"https://pith.science/pith/355I2WDQVG6H4AGLMZJ7MKX76B.json","graph_json":"https://pith.science/api/pith-number/355I2WDQVG6H4AGLMZJ7MKX76B/graph.json","events_json":"https://pith.science/api/pith-number/355I2WDQVG6H4AGLMZJ7MKX76B/events.json","paper":"https://pith.science/paper/355I2WDQ"},"agent_actions":{"view_html":"https://pith.science/pith/355I2WDQVG6H4AGLMZJ7MKX76B","download_json":"https://pith.science/pith/355I2WDQVG6H4AGLMZJ7MKX76B.json","view_paper":"https://pith.science/paper/355I2WDQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.10764&json=true","fetch_graph":"https://pith.science/api/pith-number/355I2WDQVG6H4AGLMZJ7MKX76B/graph.json","fetch_events":"https://pith.science/api/pith-number/355I2WDQVG6H4AGLMZJ7MKX76B/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/355I2WDQVG6H4AGLMZJ7MKX76B/action/timestamp_anchor","attest_storage":"https://pith.science/pith/355I2WDQVG6H4AGLMZJ7MKX76B/action/storage_attestation","attest_author":"https://pith.science/pith/355I2WDQVG6H4AGLMZJ7MKX76B/action/author_attestation","sign_citation":"https://pith.science/pith/355I2WDQVG6H4AGLMZJ7MKX76B/action/citation_signature","submit_replication":"https://pith.science/pith/355I2WDQVG6H4AGLMZJ7MKX76B/action/replication_record"}},"created_at":"2026-05-26T01:03:33.238879+00:00","updated_at":"2026-05-26T01:03:33.238879+00:00"}