{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:OBVJMSXQ5X4B5KMW46VXH4VMRV","short_pith_number":"pith:OBVJMSXQ","canonical_record":{"source":{"id":"2506.10949","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-06-12T17:50:58Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"65b9b65c910a34d32c664635b382fb6db361876b000a3b6f9c48f043374d5f17","abstract_canon_sha256":"10a06ed6cdbdd479013dee9c198c6a5de80bfad56750f63e44140ec97ead6ef8"},"schema_version":"1.0"},"canonical_sha256":"706a964af0edf81ea996e7ab73f2ac8d721969bc0d85d6c7c9f90a77c8204134","source":{"kind":"arxiv","id":"2506.10949","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.10949","created_at":"2026-07-05T11:21:49Z"},{"alias_kind":"arxiv_version","alias_value":"2506.10949v2","created_at":"2026-07-05T11:21:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.10949","created_at":"2026-07-05T11:21:49Z"},{"alias_kind":"pith_short_12","alias_value":"OBVJMSXQ5X4B","created_at":"2026-07-05T11:21:49Z"},{"alias_kind":"pith_short_16","alias_value":"OBVJMSXQ5X4B5KMW","created_at":"2026-07-05T11:21:49Z"},{"alias_kind":"pith_short_8","alias_value":"OBVJMSXQ","created_at":"2026-07-05T11:21:49Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:OBVJMSXQ5X4B5KMW46VXH4VMRV","target":"record","payload":{"canonical_record":{"source":{"id":"2506.10949","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-06-12T17:50:58Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"65b9b65c910a34d32c664635b382fb6db361876b000a3b6f9c48f043374d5f17","abstract_canon_sha256":"10a06ed6cdbdd479013dee9c198c6a5de80bfad56750f63e44140ec97ead6ef8"},"schema_version":"1.0"},"canonical_sha256":"706a964af0edf81ea996e7ab73f2ac8d721969bc0d85d6c7c9f90a77c8204134","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:21:49.810566Z","signature_b64":"ZbTmkX3K7BywRIS+g84JshKVawBEJm/WYvxgDJZRd11Lt0w0kxg4ydNiCzYK9JjQqt/hH3QtXaR4C2ZF+NQ5CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"706a964af0edf81ea996e7ab73f2ac8d721969bc0d85d6c7c9f90a77c8204134","last_reissued_at":"2026-07-05T11:21:49.810126Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:21:49.810126Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.10949","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:21:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dwbvuUivO5S2Nr00USPmSFWqCVUUVZZ7GcYyrvjbifiYuLl/r5OKVAL4uP4NSrCfTgrQxIER/oyG20JbwcY7Bg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T06:25:27.231517Z"},"content_sha256":"232ea320c27edabfbcf7c41beffbec988c1a9aa908e77d384b57d8957b7087d7","schema_version":"1.0","event_id":"sha256:232ea320c27edabfbcf7c41beffbec988c1a9aa908e77d384b57d8957b7087d7"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:OBVJMSXQ5X4B5KMW46VXH4VMRV","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Monitoring Decomposition Attacks in LLMs with Lightweight Sequential Monitors","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Chen Yueh-Han, He He, Maksym Andriushchenko, Nitish Joshi, Rico Angell, Yulin Chen","submitted_at":"2025-06-12T17:50:58Z","abstract_excerpt":"Current LLM safety defenses fail under decomposition attacks, where a malicious goal is decomposed into benign subtasks that circumvent refusals. The challenge lies in the existing shallow safety alignment techniques: they only detect harm in the immediate prompt and do not reason about long-range intent, leaving them blind to malicious intent that emerges over a sequence of seemingly benign instructions. We therefore propose adding an external monitor that observes the conversation at a higher granularity. To facilitate our study of monitoring decomposition attacks, we curate the largest and "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.10949","kind":"arxiv","version":2},"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/2506.10949/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":"2026-07-05T11:21:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"hPpKsKemNbd6f8NUoSzU0QvLYx18zeOmqVbwA2+2fEig5wlfLivPwWQ9I7iQADu5kmOHLz9Rq1vtDDZ7LnOMCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T06:25:27.231891Z"},"content_sha256":"3d39515167b81a2acefb7fd2b54e9596a2a14389a77b6c7b5f2ae8dd1a6b619c","schema_version":"1.0","event_id":"sha256:3d39515167b81a2acefb7fd2b54e9596a2a14389a77b6c7b5f2ae8dd1a6b619c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/OBVJMSXQ5X4B5KMW46VXH4VMRV/bundle.json","state_url":"https://pith.science/pith/OBVJMSXQ5X4B5KMW46VXH4VMRV/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/OBVJMSXQ5X4B5KMW46VXH4VMRV/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":"2026-07-06T06:25:27Z","links":{"resolver":"https://pith.science/pith/OBVJMSXQ5X4B5KMW46VXH4VMRV","bundle":"https://pith.science/pith/OBVJMSXQ5X4B5KMW46VXH4VMRV/bundle.json","state":"https://pith.science/pith/OBVJMSXQ5X4B5KMW46VXH4VMRV/state.json","well_known_bundle":"https://pith.science/.well-known/pith/OBVJMSXQ5X4B5KMW46VXH4VMRV/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:OBVJMSXQ5X4B5KMW46VXH4VMRV","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":"10a06ed6cdbdd479013dee9c198c6a5de80bfad56750f63e44140ec97ead6ef8","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-06-12T17:50:58Z","title_canon_sha256":"65b9b65c910a34d32c664635b382fb6db361876b000a3b6f9c48f043374d5f17"},"schema_version":"1.0","source":{"id":"2506.10949","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.10949","created_at":"2026-07-05T11:21:49Z"},{"alias_kind":"arxiv_version","alias_value":"2506.10949v2","created_at":"2026-07-05T11:21:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.10949","created_at":"2026-07-05T11:21:49Z"},{"alias_kind":"pith_short_12","alias_value":"OBVJMSXQ5X4B","created_at":"2026-07-05T11:21:49Z"},{"alias_kind":"pith_short_16","alias_value":"OBVJMSXQ5X4B5KMW","created_at":"2026-07-05T11:21:49Z"},{"alias_kind":"pith_short_8","alias_value":"OBVJMSXQ","created_at":"2026-07-05T11:21:49Z"}],"graph_snapshots":[{"event_id":"sha256:3d39515167b81a2acefb7fd2b54e9596a2a14389a77b6c7b5f2ae8dd1a6b619c","target":"graph","created_at":"2026-07-05T11:21:49Z","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/2506.10949/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Current LLM safety defenses fail under decomposition attacks, where a malicious goal is decomposed into benign subtasks that circumvent refusals. The challenge lies in the existing shallow safety alignment techniques: they only detect harm in the immediate prompt and do not reason about long-range intent, leaving them blind to malicious intent that emerges over a sequence of seemingly benign instructions. We therefore propose adding an external monitor that observes the conversation at a higher granularity. To facilitate our study of monitoring decomposition attacks, we curate the largest and ","authors_text":"Chen Yueh-Han, He He, Maksym Andriushchenko, Nitish Joshi, Rico Angell, Yulin Chen","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-06-12T17:50:58Z","title":"Monitoring Decomposition Attacks in LLMs with Lightweight Sequential Monitors"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.10949","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":"sha256:232ea320c27edabfbcf7c41beffbec988c1a9aa908e77d384b57d8957b7087d7","target":"record","created_at":"2026-07-05T11:21:49Z","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":"10a06ed6cdbdd479013dee9c198c6a5de80bfad56750f63e44140ec97ead6ef8","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-06-12T17:50:58Z","title_canon_sha256":"65b9b65c910a34d32c664635b382fb6db361876b000a3b6f9c48f043374d5f17"},"schema_version":"1.0","source":{"id":"2506.10949","kind":"arxiv","version":2}},"canonical_sha256":"706a964af0edf81ea996e7ab73f2ac8d721969bc0d85d6c7c9f90a77c8204134","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"706a964af0edf81ea996e7ab73f2ac8d721969bc0d85d6c7c9f90a77c8204134","first_computed_at":"2026-07-05T11:21:49.810126Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:21:49.810126Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ZbTmkX3K7BywRIS+g84JshKVawBEJm/WYvxgDJZRd11Lt0w0kxg4ydNiCzYK9JjQqt/hH3QtXaR4C2ZF+NQ5CA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:21:49.810566Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.10949","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:232ea320c27edabfbcf7c41beffbec988c1a9aa908e77d384b57d8957b7087d7","sha256:3d39515167b81a2acefb7fd2b54e9596a2a14389a77b6c7b5f2ae8dd1a6b619c"],"state_sha256":"7cb2811a3d6c5c050410ebc738653fef411f4163c3b69f163f263763851b7da5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lGSEW8GITgzwnekyPem2Hi6JnrQgRILVlyJ43nF+pFrCQ5SMzdeLJaqAPCu1ch/o9/aC3RSsaFkWJd97VvAPCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-06T06:25:27.233925Z","bundle_sha256":"e2109e2130a33a1e5eb3bc270c9d3a6b25ef8c2e235779b0036c0a8adb546c95"}}