{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:Y4N2MKYRLD3Z24H5JP73YPBD3Q","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":"4efc33224f361e533ab16ab783da028153096840b996d1e4ec559a88b3b0d6b5","cross_cats_sorted":["cs.AI","cs.CL","cs.CR","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-27T04:04:42Z","title_canon_sha256":"fdf5b01d8c84ed48688dbca81b423a78ae7f1d08d553adedc3c85aea000da195"},"schema_version":"1.0","source":{"id":"2605.27932","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.27932","created_at":"2026-05-28T01:04:52Z"},{"alias_kind":"arxiv_version","alias_value":"2605.27932v1","created_at":"2026-05-28T01:04:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.27932","created_at":"2026-05-28T01:04:52Z"},{"alias_kind":"pith_short_12","alias_value":"Y4N2MKYRLD3Z","created_at":"2026-05-28T01:04:52Z"},{"alias_kind":"pith_short_16","alias_value":"Y4N2MKYRLD3Z24H5","created_at":"2026-05-28T01:04:52Z"},{"alias_kind":"pith_short_8","alias_value":"Y4N2MKYR","created_at":"2026-05-28T01:04:52Z"}],"graph_snapshots":[{"event_id":"sha256:6c2e229a0408db5319795f4361da9420b6f1f19f08af17a06487a27eded3ff81","target":"graph","created_at":"2026-05-28T01:04:52Z","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/2605.27932/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Think-with-image reasoning is emerging as a new inference paradigm for large vision-language models, but its safety implications remain poorly understood. Existing systems already span multiple process designs, including direct response generation, text-only prior turn, visual-state manipulation, and explicit external image-tool invocation. In this paper, we ask which of these evaluated paradigms improves multimodal jailbreak robustness, and why. Across multiple vision-language models, explicit image-tool interaction yields the lowest attack success rates in our experiments, reducing jailbreak","authors_text":"Binghang Lu, Bing Hu, Fang Wu, Neil Zhenqiang Gong, Xiaomin Li, Yuan Tian","cross_cats":["cs.AI","cs.CL","cs.CR","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-27T04:04:42Z","title":"When Think-with-Image Meets Safety: What Determines Multimodal Jailbreak Robustness?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.27932","kind":"arxiv","version":1},"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":"sha256:227f67297d8f89817f09d0e56ad6fab9bd45c1fbbabff1fae2f2fe63b2fc6558","target":"record","created_at":"2026-05-28T01:04:52Z","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":"4efc33224f361e533ab16ab783da028153096840b996d1e4ec559a88b3b0d6b5","cross_cats_sorted":["cs.AI","cs.CL","cs.CR","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-27T04:04:42Z","title_canon_sha256":"fdf5b01d8c84ed48688dbca81b423a78ae7f1d08d553adedc3c85aea000da195"},"schema_version":"1.0","source":{"id":"2605.27932","kind":"arxiv","version":1}},"canonical_sha256":"c71ba62b1158f79d70fd4bffbc3c23dc10a2189e85ab2ed9b5042499457b42f9","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c71ba62b1158f79d70fd4bffbc3c23dc10a2189e85ab2ed9b5042499457b42f9","first_computed_at":"2026-05-28T01:04:52.937926Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-28T01:04:52.937926Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Tdk4V57J7Lf6jRf6pl9SKEEnz5DULD4ThhfQEwYQ+tmoSCO12HnXSx25cO7QwcIQsDGAYY8+OCdY9FJAApr9BQ==","signature_status":"signed_v1","signed_at":"2026-05-28T01:04:52.938349Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.27932","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:227f67297d8f89817f09d0e56ad6fab9bd45c1fbbabff1fae2f2fe63b2fc6558","sha256:6c2e229a0408db5319795f4361da9420b6f1f19f08af17a06487a27eded3ff81"],"state_sha256":"3bea48722af19b0594f93209a6f080f8fea5bf3d25207c76eb1653584d99e64a"}