{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:Q5FY5XLKXERNZWCM6WDCM63666","short_pith_number":"pith:Q5FY5XLK","canonical_record":{"source":{"id":"2405.09055","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-15T03:04:05Z","cross_cats_sorted":[],"title_canon_sha256":"2f3a477b1f783c09ba2509ed3f2493358b719366bad964e23079cc0dfd5b5f10","abstract_canon_sha256":"0eb13b6d12a09de4b782895f3449590cf2549ab1d1024ea1c1a305549b4b4478"},"schema_version":"1.0"},"canonical_sha256":"874b8edd6ab922dcd84cf586267b7ef7b9d37903a365df12a3f33c587c021fd8","source":{"kind":"arxiv","id":"2405.09055","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.09055","created_at":"2026-07-05T08:19:12Z"},{"alias_kind":"arxiv_version","alias_value":"2405.09055v1","created_at":"2026-07-05T08:19:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.09055","created_at":"2026-07-05T08:19:12Z"},{"alias_kind":"pith_short_12","alias_value":"Q5FY5XLKXERN","created_at":"2026-07-05T08:19:12Z"},{"alias_kind":"pith_short_16","alias_value":"Q5FY5XLKXERNZWCM","created_at":"2026-07-05T08:19:12Z"},{"alias_kind":"pith_short_8","alias_value":"Q5FY5XLK","created_at":"2026-07-05T08:19:12Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:Q5FY5XLKXERNZWCM6WDCM63666","target":"record","payload":{"canonical_record":{"source":{"id":"2405.09055","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-15T03:04:05Z","cross_cats_sorted":[],"title_canon_sha256":"2f3a477b1f783c09ba2509ed3f2493358b719366bad964e23079cc0dfd5b5f10","abstract_canon_sha256":"0eb13b6d12a09de4b782895f3449590cf2549ab1d1024ea1c1a305549b4b4478"},"schema_version":"1.0"},"canonical_sha256":"874b8edd6ab922dcd84cf586267b7ef7b9d37903a365df12a3f33c587c021fd8","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:19:12.512106Z","signature_b64":"7bBW5d3e5zfd4OwT7xNqE9uZxRIW7JEsL9oQI89MuVNPYyfTVZSaSux8u7pQqe8Ig+OU9HBKyFo2QhH/3z4BCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"874b8edd6ab922dcd84cf586267b7ef7b9d37903a365df12a3f33c587c021fd8","last_reissued_at":"2026-07-05T08:19:12.511695Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:19:12.511695Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2405.09055","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":"2026-07-05T08:19:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"U0kx4+dREsN5QsxFNr3VbOd//HGRfTeJ90COoO0GnztCbYtTmxNchVg/fJUXP6RYE4tH4LMlQshQiOMg5rsrCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T13:25:13.515112Z"},"content_sha256":"025ca918c717ee3103d61f4b7dc5b804f108a2c623f8ed29d83536850f3a2932","schema_version":"1.0","event_id":"sha256:025ca918c717ee3103d61f4b7dc5b804f108a2c623f8ed29d83536850f3a2932"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:Q5FY5XLKXERNZWCM6WDCM63666","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"A safety realignment framework via subspace-oriented model fusion for large language models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Liang He, Linlin Wang, Shunfan Zheng, Xiaoling Wang, Xin Yi","submitted_at":"2024-05-15T03:04:05Z","abstract_excerpt":"The current safeguard mechanisms for large language models (LLMs) are indeed susceptible to jailbreak attacks, making them inherently fragile. Even the process of fine-tuning on apparently benign data for downstream tasks can jeopardize safety. One potential solution is to conduct safety fine-tuning subsequent to downstream fine-tuning. However, there's a risk of catastrophic forgetting during safety fine-tuning, where LLMs may regain safety measures but lose the task-specific knowledge acquired during downstream fine-tuning. In this paper, we introduce a safety realignment framework through s"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.09055","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/2405.09055/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":"2026-07-05T08:19:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pOWNNiKqGUABTrpMcvD7Hzi2cGlVb51eC7C2cKK8p4PbauNzi1R3TRIqQLylYdjCAD58I7x6/1AvPqwseK4yAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T13:25:13.515966Z"},"content_sha256":"5b3950e1fa8f4035ef498b1567926875c829723863a2aa3aa39117adb8d95fa8","schema_version":"1.0","event_id":"sha256:5b3950e1fa8f4035ef498b1567926875c829723863a2aa3aa39117adb8d95fa8"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/Q5FY5XLKXERNZWCM6WDCM63666/bundle.json","state_url":"https://pith.science/pith/Q5FY5XLKXERNZWCM6WDCM63666/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/Q5FY5XLKXERNZWCM6WDCM63666/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":"2026-08-09T13:25:13Z","links":{"resolver":"https://pith.science/pith/Q5FY5XLKXERNZWCM6WDCM63666","bundle":"https://pith.science/pith/Q5FY5XLKXERNZWCM6WDCM63666/bundle.json","state":"https://pith.science/pith/Q5FY5XLKXERNZWCM6WDCM63666/state.json","well_known_bundle":"https://pith.science/.well-known/pith/Q5FY5XLKXERNZWCM6WDCM63666/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:Q5FY5XLKXERNZWCM6WDCM63666","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":"0eb13b6d12a09de4b782895f3449590cf2549ab1d1024ea1c1a305549b4b4478","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-15T03:04:05Z","title_canon_sha256":"2f3a477b1f783c09ba2509ed3f2493358b719366bad964e23079cc0dfd5b5f10"},"schema_version":"1.0","source":{"id":"2405.09055","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.09055","created_at":"2026-07-05T08:19:12Z"},{"alias_kind":"arxiv_version","alias_value":"2405.09055v1","created_at":"2026-07-05T08:19:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.09055","created_at":"2026-07-05T08:19:12Z"},{"alias_kind":"pith_short_12","alias_value":"Q5FY5XLKXERN","created_at":"2026-07-05T08:19:12Z"},{"alias_kind":"pith_short_16","alias_value":"Q5FY5XLKXERNZWCM","created_at":"2026-07-05T08:19:12Z"},{"alias_kind":"pith_short_8","alias_value":"Q5FY5XLK","created_at":"2026-07-05T08:19:12Z"}],"graph_snapshots":[{"event_id":"sha256:5b3950e1fa8f4035ef498b1567926875c829723863a2aa3aa39117adb8d95fa8","target":"graph","created_at":"2026-07-05T08:19:12Z","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/2405.09055/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The current safeguard mechanisms for large language models (LLMs) are indeed susceptible to jailbreak attacks, making them inherently fragile. Even the process of fine-tuning on apparently benign data for downstream tasks can jeopardize safety. One potential solution is to conduct safety fine-tuning subsequent to downstream fine-tuning. However, there's a risk of catastrophic forgetting during safety fine-tuning, where LLMs may regain safety measures but lose the task-specific knowledge acquired during downstream fine-tuning. In this paper, we introduce a safety realignment framework through s","authors_text":"Liang He, Linlin Wang, Shunfan Zheng, Xiaoling Wang, Xin Yi","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-15T03:04:05Z","title":"A safety realignment framework via subspace-oriented model fusion for large language models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.09055","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":"sha256:025ca918c717ee3103d61f4b7dc5b804f108a2c623f8ed29d83536850f3a2932","target":"record","created_at":"2026-07-05T08:19:12Z","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":"0eb13b6d12a09de4b782895f3449590cf2549ab1d1024ea1c1a305549b4b4478","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-15T03:04:05Z","title_canon_sha256":"2f3a477b1f783c09ba2509ed3f2493358b719366bad964e23079cc0dfd5b5f10"},"schema_version":"1.0","source":{"id":"2405.09055","kind":"arxiv","version":1}},"canonical_sha256":"874b8edd6ab922dcd84cf586267b7ef7b9d37903a365df12a3f33c587c021fd8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"874b8edd6ab922dcd84cf586267b7ef7b9d37903a365df12a3f33c587c021fd8","first_computed_at":"2026-07-05T08:19:12.511695Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:19:12.511695Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"7bBW5d3e5zfd4OwT7xNqE9uZxRIW7JEsL9oQI89MuVNPYyfTVZSaSux8u7pQqe8Ig+OU9HBKyFo2QhH/3z4BCA==","signature_status":"signed_v1","signed_at":"2026-07-05T08:19:12.512106Z","signed_message":"canonical_sha256_bytes"},"source_id":"2405.09055","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:025ca918c717ee3103d61f4b7dc5b804f108a2c623f8ed29d83536850f3a2932","sha256:5b3950e1fa8f4035ef498b1567926875c829723863a2aa3aa39117adb8d95fa8"],"state_sha256":"6b65a39d13187679c8cb5c9d17938ed3779a11d4b9144fdd6dd8d96849503e19"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"iIuPv9hSxnX6jgByXUg3l/hWH9NPykslwp/u6qruY2owQPQaGStvCx9Otq1DXAUSefVKWvIRBmKaXglIWjG8CA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T13:25:13.522605Z","bundle_sha256":"84040c76ba475ce44fe78ec2c30975f8eb7c3ab4f9fbd3bb2c71a728ac381805"}}