{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:UBFUKC2Y7357I3TFO4T5E2BZIS","short_pith_number":"pith:UBFUKC2Y","canonical_record":{"source":{"id":"2502.14744","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-20T17:14:34Z","cross_cats_sorted":[],"title_canon_sha256":"a9b4c048e7df139f9ae2eb381ca54f59e8070476da50ab3227ed4c8aca92b2ed","abstract_canon_sha256":"db5eb6f04253c05be7a1b322a5590c2a8744b1cae8e1bc5cf1f0ebdf0e50bfae"},"schema_version":"1.0"},"canonical_sha256":"a04b450b58fefbf46e657727d2683944b1d63e9759971b6a8ef3fe82eb0866d4","source":{"kind":"arxiv","id":"2502.14744","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.14744","created_at":"2026-07-05T11:25:42Z"},{"alias_kind":"arxiv_version","alias_value":"2502.14744v4","created_at":"2026-07-05T11:25:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.14744","created_at":"2026-07-05T11:25:42Z"},{"alias_kind":"pith_short_12","alias_value":"UBFUKC2Y7357","created_at":"2026-07-05T11:25:42Z"},{"alias_kind":"pith_short_16","alias_value":"UBFUKC2Y7357I3TF","created_at":"2026-07-05T11:25:42Z"},{"alias_kind":"pith_short_8","alias_value":"UBFUKC2Y","created_at":"2026-07-05T11:25:42Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:UBFUKC2Y7357I3TFO4T5E2BZIS","target":"record","payload":{"canonical_record":{"source":{"id":"2502.14744","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-20T17:14:34Z","cross_cats_sorted":[],"title_canon_sha256":"a9b4c048e7df139f9ae2eb381ca54f59e8070476da50ab3227ed4c8aca92b2ed","abstract_canon_sha256":"db5eb6f04253c05be7a1b322a5590c2a8744b1cae8e1bc5cf1f0ebdf0e50bfae"},"schema_version":"1.0"},"canonical_sha256":"a04b450b58fefbf46e657727d2683944b1d63e9759971b6a8ef3fe82eb0866d4","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:25:42.249074Z","signature_b64":"ym3NcRj0h5ps3gJySIaSX3EUSe8hMJzAO9fFK8eYH1ehAQv8ytwU4cqX3zWBIYp/iQnAeu6FSlt+7WcfNHB0Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a04b450b58fefbf46e657727d2683944b1d63e9759971b6a8ef3fe82eb0866d4","last_reissued_at":"2026-07-05T11:25:42.248591Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:25:42.248591Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2502.14744","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:25:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"azScNmApZycQh/C6gbE3N72a0eKED1+nv0aMRwBzcxDiPduKNU9EQluvOdIHGNfpR0n45lNON7IqR4hi9liIBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T12:41:15.062168Z"},"content_sha256":"38eb3672d15655ed7204c45f05285ef897d72674ebb543200161d95f66ae128d","schema_version":"1.0","event_id":"sha256:38eb3672d15655ed7204c45f05285ef897d72674ebb543200161d95f66ae128d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:UBFUKC2Y7357I3TFO4T5E2BZIS","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"HiddenDetect: Detecting Jailbreak Attacks against Large Vision-Language Models via Monitoring Hidden States","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bo Zheng, Tianshuo Peng, Xiangyu Yue, Xiaoyong Zhu, Xinyan Gao, Yilei Jiang, Yingshui Tan","submitted_at":"2025-02-20T17:14:34Z","abstract_excerpt":"The integration of additional modalities increases the susceptibility of large vision-language models (LVLMs) to safety risks, such as jailbreak attacks, compared to their language-only counterparts. While existing research primarily focuses on post-hoc alignment techniques, the underlying safety mechanisms within LVLMs remain largely unexplored. In this work , we investigate whether LVLMs inherently encode safety-relevant signals within their internal activations during inference. Our findings reveal that LVLMs exhibit distinct activation patterns when processing unsafe prompts, which can be "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.14744","kind":"arxiv","version":4},"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/2502.14744/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":"2026-07-05T11:25:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nJVX3sq1jnN1vTebWdIyzuVR8PeRa+pyuNMAWFCGCGqiRDczp/sVZam+qoAh6qhpLqYmgHzbNdOPxwFlxwd7Bg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T12:41:15.062697Z"},"content_sha256":"5a56db5752028182fea9bf92afd14d23897f0e9d3a45ca1ae8ddf3ba3edf29be","schema_version":"1.0","event_id":"sha256:5a56db5752028182fea9bf92afd14d23897f0e9d3a45ca1ae8ddf3ba3edf29be"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/UBFUKC2Y7357I3TFO4T5E2BZIS/bundle.json","state_url":"https://pith.science/pith/UBFUKC2Y7357I3TFO4T5E2BZIS/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/UBFUKC2Y7357I3TFO4T5E2BZIS/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":"2026-08-04T12:41:15Z","links":{"resolver":"https://pith.science/pith/UBFUKC2Y7357I3TFO4T5E2BZIS","bundle":"https://pith.science/pith/UBFUKC2Y7357I3TFO4T5E2BZIS/bundle.json","state":"https://pith.science/pith/UBFUKC2Y7357I3TFO4T5E2BZIS/state.json","well_known_bundle":"https://pith.science/.well-known/pith/UBFUKC2Y7357I3TFO4T5E2BZIS/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:UBFUKC2Y7357I3TFO4T5E2BZIS","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":"db5eb6f04253c05be7a1b322a5590c2a8744b1cae8e1bc5cf1f0ebdf0e50bfae","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-20T17:14:34Z","title_canon_sha256":"a9b4c048e7df139f9ae2eb381ca54f59e8070476da50ab3227ed4c8aca92b2ed"},"schema_version":"1.0","source":{"id":"2502.14744","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.14744","created_at":"2026-07-05T11:25:42Z"},{"alias_kind":"arxiv_version","alias_value":"2502.14744v4","created_at":"2026-07-05T11:25:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.14744","created_at":"2026-07-05T11:25:42Z"},{"alias_kind":"pith_short_12","alias_value":"UBFUKC2Y7357","created_at":"2026-07-05T11:25:42Z"},{"alias_kind":"pith_short_16","alias_value":"UBFUKC2Y7357I3TF","created_at":"2026-07-05T11:25:42Z"},{"alias_kind":"pith_short_8","alias_value":"UBFUKC2Y","created_at":"2026-07-05T11:25:42Z"}],"graph_snapshots":[{"event_id":"sha256:5a56db5752028182fea9bf92afd14d23897f0e9d3a45ca1ae8ddf3ba3edf29be","target":"graph","created_at":"2026-07-05T11:25:42Z","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/2502.14744/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The integration of additional modalities increases the susceptibility of large vision-language models (LVLMs) to safety risks, such as jailbreak attacks, compared to their language-only counterparts. While existing research primarily focuses on post-hoc alignment techniques, the underlying safety mechanisms within LVLMs remain largely unexplored. In this work , we investigate whether LVLMs inherently encode safety-relevant signals within their internal activations during inference. Our findings reveal that LVLMs exhibit distinct activation patterns when processing unsafe prompts, which can be ","authors_text":"Bo Zheng, Tianshuo Peng, Xiangyu Yue, Xiaoyong Zhu, Xinyan Gao, Yilei Jiang, Yingshui Tan","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-20T17:14:34Z","title":"HiddenDetect: Detecting Jailbreak Attacks against Large Vision-Language Models via Monitoring Hidden States"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.14744","kind":"arxiv","version":4},"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":"sha256:38eb3672d15655ed7204c45f05285ef897d72674ebb543200161d95f66ae128d","target":"record","created_at":"2026-07-05T11:25:42Z","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":"db5eb6f04253c05be7a1b322a5590c2a8744b1cae8e1bc5cf1f0ebdf0e50bfae","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-20T17:14:34Z","title_canon_sha256":"a9b4c048e7df139f9ae2eb381ca54f59e8070476da50ab3227ed4c8aca92b2ed"},"schema_version":"1.0","source":{"id":"2502.14744","kind":"arxiv","version":4}},"canonical_sha256":"a04b450b58fefbf46e657727d2683944b1d63e9759971b6a8ef3fe82eb0866d4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a04b450b58fefbf46e657727d2683944b1d63e9759971b6a8ef3fe82eb0866d4","first_computed_at":"2026-07-05T11:25:42.248591Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:25:42.248591Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ym3NcRj0h5ps3gJySIaSX3EUSe8hMJzAO9fFK8eYH1ehAQv8ytwU4cqX3zWBIYp/iQnAeu6FSlt+7WcfNHB0Cw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:25:42.249074Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.14744","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:38eb3672d15655ed7204c45f05285ef897d72674ebb543200161d95f66ae128d","sha256:5a56db5752028182fea9bf92afd14d23897f0e9d3a45ca1ae8ddf3ba3edf29be"],"state_sha256":"6293235b12986e5cd220bff614b3517197e6aacf381a1e28dfd0cde56fa6b3f7"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kNgmtGo4O/gPY2sKiDUCp0YBJ30XABWV896VygLoZKSIxy+vhAZ1lJ0955l15WZjyr4Wg23tO+vN+7lG86ymDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T12:41:15.067559Z","bundle_sha256":"a56c413d4e53bd354ad0bdb77653a05032572a60eb465a39e03744fb95269da4"}}