{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:CUGIL2OTP2FNC2XVTNBOTPZFUT","short_pith_number":"pith:CUGIL2OT","canonical_record":{"source":{"id":"2406.13294","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.MM","submitted_at":"2024-06-19T07:32:55Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"8cc09b9164df15103375349ee4c3f84ef827506a4e7089694649879488babc92","abstract_canon_sha256":"80d8665765f390b2e9444a3d3e950e9f0f7355a2d3c585690c5b7d3325372f48"},"schema_version":"1.0"},"canonical_sha256":"150c85e9d37e8ad16af59b42e9bf25a4e997c8e263bcef102bf8c3be200ca1ab","source":{"kind":"arxiv","id":"2406.13294","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.13294","created_at":"2026-07-05T08:34:27Z"},{"alias_kind":"arxiv_version","alias_value":"2406.13294v1","created_at":"2026-07-05T08:34:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.13294","created_at":"2026-07-05T08:34:27Z"},{"alias_kind":"pith_short_12","alias_value":"CUGIL2OTP2FN","created_at":"2026-07-05T08:34:27Z"},{"alias_kind":"pith_short_16","alias_value":"CUGIL2OTP2FNC2XV","created_at":"2026-07-05T08:34:27Z"},{"alias_kind":"pith_short_8","alias_value":"CUGIL2OT","created_at":"2026-07-05T08:34:27Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:CUGIL2OTP2FNC2XVTNBOTPZFUT","target":"record","payload":{"canonical_record":{"source":{"id":"2406.13294","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.MM","submitted_at":"2024-06-19T07:32:55Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"8cc09b9164df15103375349ee4c3f84ef827506a4e7089694649879488babc92","abstract_canon_sha256":"80d8665765f390b2e9444a3d3e950e9f0f7355a2d3c585690c5b7d3325372f48"},"schema_version":"1.0"},"canonical_sha256":"150c85e9d37e8ad16af59b42e9bf25a4e997c8e263bcef102bf8c3be200ca1ab","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:34:27.978954Z","signature_b64":"S177OxEGjctIPEsTA1NUsS46IYY5xDJPwgn6ycCgn5S54RoQhr2pufSV14VuYj8lKH8lO5CUuV802WyoOMLaBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"150c85e9d37e8ad16af59b42e9bf25a4e997c8e263bcef102bf8c3be200ca1ab","last_reissued_at":"2026-07-05T08:34:27.978403Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:34:27.978403Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2406.13294","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:34:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"cgejl0nF6cBz/BgSBJ640woMRwY9g/HH10UGL9F0TovRVyjjGVFfnYVtkeoNckC79xLTRmytUkMlnVfVaKqWDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T06:46:15.530962Z"},"content_sha256":"2770160785eb92c66f3e4f39ed39546e328d0dfa06e5ff71b212f66cf83da4f9","schema_version":"1.0","event_id":"sha256:2770160785eb92c66f3e4f39ed39546e328d0dfa06e5ff71b212f66cf83da4f9"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:CUGIL2OTP2FNC2XVTNBOTPZFUT","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Enhancing Cross-Prompt Transferability in Vision-Language Models through Contextual Injection of Target Tokens","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.MM","authors_text":"Fuqing Zhu, Jizhong Han, Songlin Hu, Xikang Yang, Xuehai Tang","submitted_at":"2024-06-19T07:32:55Z","abstract_excerpt":"Vision-language models (VLMs) seamlessly integrate visual and textual data to perform tasks such as image classification, caption generation, and visual question answering. However, adversarial images often struggle to deceive all prompts effectively in the context of cross-prompt migration attacks, as the probability distribution of the tokens in these images tends to favor the semantics of the original image rather than the target tokens. To address this challenge, we propose a Contextual-Injection Attack (CIA) that employs gradient-based perturbation to inject target tokens into both visual"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.13294","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/2406.13294/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:34:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"fR87yiLHH/8tni63JdSP5RmwkisEd19NxEwvwFLSVNKiV8BOteMNdtOJp4dVb6GnNM4cSXXp8AEMRxR/gZetDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T06:46:15.531554Z"},"content_sha256":"c1be2740fefe7b907149ac37c99a6b4da87793bffe1b97e27a3fa1f31f75e671","schema_version":"1.0","event_id":"sha256:c1be2740fefe7b907149ac37c99a6b4da87793bffe1b97e27a3fa1f31f75e671"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/CUGIL2OTP2FNC2XVTNBOTPZFUT/bundle.json","state_url":"https://pith.science/pith/CUGIL2OTP2FNC2XVTNBOTPZFUT/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/CUGIL2OTP2FNC2XVTNBOTPZFUT/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-09T06:46:15Z","links":{"resolver":"https://pith.science/pith/CUGIL2OTP2FNC2XVTNBOTPZFUT","bundle":"https://pith.science/pith/CUGIL2OTP2FNC2XVTNBOTPZFUT/bundle.json","state":"https://pith.science/pith/CUGIL2OTP2FNC2XVTNBOTPZFUT/state.json","well_known_bundle":"https://pith.science/.well-known/pith/CUGIL2OTP2FNC2XVTNBOTPZFUT/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:CUGIL2OTP2FNC2XVTNBOTPZFUT","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":"80d8665765f390b2e9444a3d3e950e9f0f7355a2d3c585690c5b7d3325372f48","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.MM","submitted_at":"2024-06-19T07:32:55Z","title_canon_sha256":"8cc09b9164df15103375349ee4c3f84ef827506a4e7089694649879488babc92"},"schema_version":"1.0","source":{"id":"2406.13294","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.13294","created_at":"2026-07-05T08:34:27Z"},{"alias_kind":"arxiv_version","alias_value":"2406.13294v1","created_at":"2026-07-05T08:34:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.13294","created_at":"2026-07-05T08:34:27Z"},{"alias_kind":"pith_short_12","alias_value":"CUGIL2OTP2FN","created_at":"2026-07-05T08:34:27Z"},{"alias_kind":"pith_short_16","alias_value":"CUGIL2OTP2FNC2XV","created_at":"2026-07-05T08:34:27Z"},{"alias_kind":"pith_short_8","alias_value":"CUGIL2OT","created_at":"2026-07-05T08:34:27Z"}],"graph_snapshots":[{"event_id":"sha256:c1be2740fefe7b907149ac37c99a6b4da87793bffe1b97e27a3fa1f31f75e671","target":"graph","created_at":"2026-07-05T08:34:27Z","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/2406.13294/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Vision-language models (VLMs) seamlessly integrate visual and textual data to perform tasks such as image classification, caption generation, and visual question answering. However, adversarial images often struggle to deceive all prompts effectively in the context of cross-prompt migration attacks, as the probability distribution of the tokens in these images tends to favor the semantics of the original image rather than the target tokens. To address this challenge, we propose a Contextual-Injection Attack (CIA) that employs gradient-based perturbation to inject target tokens into both visual","authors_text":"Fuqing Zhu, Jizhong Han, Songlin Hu, Xikang Yang, Xuehai Tang","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.MM","submitted_at":"2024-06-19T07:32:55Z","title":"Enhancing Cross-Prompt Transferability in Vision-Language Models through Contextual Injection of Target Tokens"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.13294","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":"sha256:2770160785eb92c66f3e4f39ed39546e328d0dfa06e5ff71b212f66cf83da4f9","target":"record","created_at":"2026-07-05T08:34:27Z","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":"80d8665765f390b2e9444a3d3e950e9f0f7355a2d3c585690c5b7d3325372f48","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.MM","submitted_at":"2024-06-19T07:32:55Z","title_canon_sha256":"8cc09b9164df15103375349ee4c3f84ef827506a4e7089694649879488babc92"},"schema_version":"1.0","source":{"id":"2406.13294","kind":"arxiv","version":1}},"canonical_sha256":"150c85e9d37e8ad16af59b42e9bf25a4e997c8e263bcef102bf8c3be200ca1ab","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"150c85e9d37e8ad16af59b42e9bf25a4e997c8e263bcef102bf8c3be200ca1ab","first_computed_at":"2026-07-05T08:34:27.978403Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:34:27.978403Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"S177OxEGjctIPEsTA1NUsS46IYY5xDJPwgn6ycCgn5S54RoQhr2pufSV14VuYj8lKH8lO5CUuV802WyoOMLaBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T08:34:27.978954Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.13294","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2770160785eb92c66f3e4f39ed39546e328d0dfa06e5ff71b212f66cf83da4f9","sha256:c1be2740fefe7b907149ac37c99a6b4da87793bffe1b97e27a3fa1f31f75e671"],"state_sha256":"ebebd14b318cc050afd54daef42955c27b8ba6123ccfc6134693aeb6adf4a55e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"accPoLbQJJ1KhieSo55WdLopPKB6SMByjzg/CnfER8B5vk0hxsTpkWZqRjPzpdpZzUx7VHPPfJZqwNbTBfHUDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T06:46:15.537149Z","bundle_sha256":"60c791432149c29fac29b5bb9fefa752550916d607521dc49f1c2a223b1eb202"}}