{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:4SZQPFNMFYPDT5UY7KJGXSTMPA","short_pith_number":"pith:4SZQPFNM","canonical_record":{"source":{"id":"2607.14147","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-14T09:30:34Z","cross_cats_sorted":["cs.AI","cs.CR","cs.LG"],"title_canon_sha256":"9bdf0b1ed0419e54a1fab33f2f5bcf79dd653047d03984737a30797ad82e7068","abstract_canon_sha256":"42b3d5d52b7ab6d49570f45b696acfa5e1f8b5971ca9b383873d98e7cd60520e"},"schema_version":"1.0"},"canonical_sha256":"e4b30795ac2e1e39f698fa926bca6c780402fc7894fc57fd6073e190c2da87af","source":{"kind":"arxiv","id":"2607.14147","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.14147","created_at":"2026-07-17T00:20:54Z"},{"alias_kind":"arxiv_version","alias_value":"2607.14147v1","created_at":"2026-07-17T00:20:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.14147","created_at":"2026-07-17T00:20:54Z"},{"alias_kind":"pith_short_12","alias_value":"4SZQPFNMFYPD","created_at":"2026-07-17T00:20:54Z"},{"alias_kind":"pith_short_16","alias_value":"4SZQPFNMFYPDT5UY","created_at":"2026-07-17T00:20:54Z"},{"alias_kind":"pith_short_8","alias_value":"4SZQPFNM","created_at":"2026-07-17T00:20:54Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:4SZQPFNMFYPDT5UY7KJGXSTMPA","target":"record","payload":{"canonical_record":{"source":{"id":"2607.14147","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-14T09:30:34Z","cross_cats_sorted":["cs.AI","cs.CR","cs.LG"],"title_canon_sha256":"9bdf0b1ed0419e54a1fab33f2f5bcf79dd653047d03984737a30797ad82e7068","abstract_canon_sha256":"42b3d5d52b7ab6d49570f45b696acfa5e1f8b5971ca9b383873d98e7cd60520e"},"schema_version":"1.0"},"canonical_sha256":"e4b30795ac2e1e39f698fa926bca6c780402fc7894fc57fd6073e190c2da87af","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-17T00:20:54.003178Z","signature_b64":"X51EoXUAgqNEq0/v/j+jI9agExTSzux4qyiup7BkQYqffSRK/1qCVwJG1d/4HJshB4L0jHBe8xZDzfjeUp40Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e4b30795ac2e1e39f698fa926bca6c780402fc7894fc57fd6073e190c2da87af","last_reissued_at":"2026-07-17T00:20:54.002328Z","signature_status":"signed_v1","first_computed_at":"2026-07-17T00:20:54.002328Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.14147","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":"2026-07-17T00:20:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wUWctQ2gpGyeV/V/xz0YiKpzs8+RjwZv6JQ70fI8q9xGnaxphBaMEMqBc8s7O4gRFfq3fD7crLH/mxduW1DKAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T22:40:22.234042Z"},"content_sha256":"cab605f8155f90db0878cd94550e2ae94d077f57ae07d3b7c444aff2177a37b8","schema_version":"1.0","event_id":"sha256:cab605f8155f90db0878cd94550e2ae94d077f57ae07d3b7c444aff2177a37b8"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:4SZQPFNMFYPDT5UY7KJGXSTMPA","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Breaking Refusal in the First Half: A Mechanistic Study of the Prefill Jailbreak","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CR","cs.LG"],"primary_cat":"cs.CL","authors_text":"Alex Kwon","submitted_at":"2026-07-14T09:30:34Z","abstract_excerpt":"Aligned language models refuse harmful requests, but a one-line prefill (\"Sure, here is\") strips the refusal. We ask where and how it fails. The harm representation stays intact: on the prompts the attack flips to compliance, a linear probe reads harm as high as on the refused ones (0.91-0.98), while behavioral refusal drops to chance. This holds across four models and three families (1.5-3.8B, and at 14B). Refusal is therefore a shallow, response-site computation. We localize it to an early window: a dose-matched position control shows the first half of the response suffices to break refusal,"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.14147","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/2607.14147/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":"2026-07-17T00:20:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Z6jsm2nZT5PFHyWkZNqcFH8yd0+lLzcxI18jGMS5jZFHf1J4mxhhMHLGVS0YLrm+VpCBk1KeB13kUIoDhI9kBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T22:40:22.234604Z"},"content_sha256":"340e218c1c1ea63a0ca8a01854dcead9a09c55c6b55febc940f31dee30f26432","schema_version":"1.0","event_id":"sha256:340e218c1c1ea63a0ca8a01854dcead9a09c55c6b55febc940f31dee30f26432"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4SZQPFNMFYPDT5UY7KJGXSTMPA/bundle.json","state_url":"https://pith.science/pith/4SZQPFNMFYPDT5UY7KJGXSTMPA/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4SZQPFNMFYPDT5UY7KJGXSTMPA/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":"2026-08-04T22:40:22Z","links":{"resolver":"https://pith.science/pith/4SZQPFNMFYPDT5UY7KJGXSTMPA","bundle":"https://pith.science/pith/4SZQPFNMFYPDT5UY7KJGXSTMPA/bundle.json","state":"https://pith.science/pith/4SZQPFNMFYPDT5UY7KJGXSTMPA/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4SZQPFNMFYPDT5UY7KJGXSTMPA/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:4SZQPFNMFYPDT5UY7KJGXSTMPA","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":"42b3d5d52b7ab6d49570f45b696acfa5e1f8b5971ca9b383873d98e7cd60520e","cross_cats_sorted":["cs.AI","cs.CR","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-14T09:30:34Z","title_canon_sha256":"9bdf0b1ed0419e54a1fab33f2f5bcf79dd653047d03984737a30797ad82e7068"},"schema_version":"1.0","source":{"id":"2607.14147","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.14147","created_at":"2026-07-17T00:20:54Z"},{"alias_kind":"arxiv_version","alias_value":"2607.14147v1","created_at":"2026-07-17T00:20:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.14147","created_at":"2026-07-17T00:20:54Z"},{"alias_kind":"pith_short_12","alias_value":"4SZQPFNMFYPD","created_at":"2026-07-17T00:20:54Z"},{"alias_kind":"pith_short_16","alias_value":"4SZQPFNMFYPDT5UY","created_at":"2026-07-17T00:20:54Z"},{"alias_kind":"pith_short_8","alias_value":"4SZQPFNM","created_at":"2026-07-17T00:20:54Z"}],"graph_snapshots":[{"event_id":"sha256:340e218c1c1ea63a0ca8a01854dcead9a09c55c6b55febc940f31dee30f26432","target":"graph","created_at":"2026-07-17T00:20:54Z","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/2607.14147/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Aligned language models refuse harmful requests, but a one-line prefill (\"Sure, here is\") strips the refusal. We ask where and how it fails. The harm representation stays intact: on the prompts the attack flips to compliance, a linear probe reads harm as high as on the refused ones (0.91-0.98), while behavioral refusal drops to chance. This holds across four models and three families (1.5-3.8B, and at 14B). Refusal is therefore a shallow, response-site computation. We localize it to an early window: a dose-matched position control shows the first half of the response suffices to break refusal,","authors_text":"Alex Kwon","cross_cats":["cs.AI","cs.CR","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-14T09:30:34Z","title":"Breaking Refusal in the First Half: A Mechanistic Study of the Prefill Jailbreak"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.14147","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:cab605f8155f90db0878cd94550e2ae94d077f57ae07d3b7c444aff2177a37b8","target":"record","created_at":"2026-07-17T00:20:54Z","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":"42b3d5d52b7ab6d49570f45b696acfa5e1f8b5971ca9b383873d98e7cd60520e","cross_cats_sorted":["cs.AI","cs.CR","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-14T09:30:34Z","title_canon_sha256":"9bdf0b1ed0419e54a1fab33f2f5bcf79dd653047d03984737a30797ad82e7068"},"schema_version":"1.0","source":{"id":"2607.14147","kind":"arxiv","version":1}},"canonical_sha256":"e4b30795ac2e1e39f698fa926bca6c780402fc7894fc57fd6073e190c2da87af","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e4b30795ac2e1e39f698fa926bca6c780402fc7894fc57fd6073e190c2da87af","first_computed_at":"2026-07-17T00:20:54.002328Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-17T00:20:54.002328Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"X51EoXUAgqNEq0/v/j+jI9agExTSzux4qyiup7BkQYqffSRK/1qCVwJG1d/4HJshB4L0jHBe8xZDzfjeUp40Bg==","signature_status":"signed_v1","signed_at":"2026-07-17T00:20:54.003178Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.14147","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:cab605f8155f90db0878cd94550e2ae94d077f57ae07d3b7c444aff2177a37b8","sha256:340e218c1c1ea63a0ca8a01854dcead9a09c55c6b55febc940f31dee30f26432"],"state_sha256":"30c43c0cfb1c8c099c20c0cde39f098fa00ba11578fa079be7fe9018b515f15b"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ZM3j9t75EDjQd6jtHCv3hs7LyCqdJ7oTSidmvRoX2p9AMhmBQmYWpgSkjf134Isxp7UfYwPjGNSBtdN6KOqXAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T22:40:22.239781Z","bundle_sha256":"cc681869f8e48f01932fa9d69a2cf946a83b6cf26d43f07070633cd6294e7ac0"}}