{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:3JGUX22ELHYZASW6HKG27GVJYW","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"ace239d7541ec24b488a14a18c39d3f66ccaa7adb747bfd5bb767054c9f938a9","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-10-28T12:43:47Z","title_canon_sha256":"466a5844ad20c0279e9134e002e3479bccbc602d47a8bcc7ebda73b6bfbec2c9"},"schema_version":"1.0","source":{"id":"2410.20971","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.20971","created_at":"2026-07-05T10:13:04Z"},{"alias_kind":"arxiv_version","alias_value":"2410.20971v2","created_at":"2026-07-05T10:13:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.20971","created_at":"2026-07-05T10:13:04Z"},{"alias_kind":"pith_short_12","alias_value":"3JGUX22ELHYZ","created_at":"2026-07-05T10:13:04Z"},{"alias_kind":"pith_short_16","alias_value":"3JGUX22ELHYZASW6","created_at":"2026-07-05T10:13:04Z"},{"alias_kind":"pith_short_8","alias_value":"3JGUX22E","created_at":"2026-07-05T10:13:04Z"}],"graph_snapshots":[{"event_id":"sha256:994d7cc0f25f9e4c304b43ba285447b9e3a21d034988b51818eb87963b564d6c","target":"graph","created_at":"2026-07-05T10:13:04Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2410.20971/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"In this paper, we focus on black-box defense for VLMs against jailbreak attacks. Existing black-box defense methods are either unimodal or bimodal. Unimodal methods enhance either the vision or language module of the VLM, while bimodal methods robustify the model through text-image representation realignment. However, these methods suffer from two limitations: 1) they fail to fully exploit the cross-modal information, or 2) they degrade the model performance on benign inputs. To address these limitations, we propose a novel blue-team method BlueSuffix that defends target VLMs against jailbreak","authors_text":"Lin Luo, Xiang Zheng, Xingjun Ma, Yige Li, Yu-Gang Jiang, Yunhan Zhao","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-10-28T12:43:47Z","title":"BlueSuffix: Reinforced Blue Teaming for Vision-Language Models Against Jailbreak Attacks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.20971","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:e90b0bc13e0a149841d795228eec5b3599480cd33a2aec5995118586061623cc","target":"record","created_at":"2026-07-05T10:13:04Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"ace239d7541ec24b488a14a18c39d3f66ccaa7adb747bfd5bb767054c9f938a9","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-10-28T12:43:47Z","title_canon_sha256":"466a5844ad20c0279e9134e002e3479bccbc602d47a8bcc7ebda73b6bfbec2c9"},"schema_version":"1.0","source":{"id":"2410.20971","kind":"arxiv","version":2}},"canonical_sha256":"da4d4beb4459f1904ade3a8daf9aa9c5bf28da622b6080dc2efba84e72ffd4d9","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"da4d4beb4459f1904ade3a8daf9aa9c5bf28da622b6080dc2efba84e72ffd4d9","first_computed_at":"2026-07-05T10:13:04.996301Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:13:04.996301Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"dBXyKvEDBYAg8HKRgMxcUZHLO6BoJk5yOUMvK/WD0hcoFHWC+xWK27siQYmeM/YjE2ZpBh+0+dh5Adj/dQrVCA==","signature_status":"signed_v1","signed_at":"2026-07-05T10:13:04.996823Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.20971","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e90b0bc13e0a149841d795228eec5b3599480cd33a2aec5995118586061623cc","sha256:994d7cc0f25f9e4c304b43ba285447b9e3a21d034988b51818eb87963b564d6c"],"state_sha256":"456639aacab2aa218a70caeddf591537b251a3464237502f73efa5d86d76e905"}