{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:YUKIMGA4UUUGPKVBUPZ7TVOSBL","short_pith_number":"pith:YUKIMGA4","canonical_record":{"source":{"id":"2605.24583","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-23T13:47:17Z","cross_cats_sorted":["cs.CL","stat.ML"],"title_canon_sha256":"90e9f16fa30048d951bde5e3340d28d44f87b49a4c5a8a8d175b98a9e6a7f3cf","abstract_canon_sha256":"a641204a06f864c11f5a8dddd05d40e6ff8a7c1324d2e44d8c4e89b13f97e1b1"},"schema_version":"1.0"},"canonical_sha256":"c51486181ca52867aaa1a3f3f9d5d20ae7d938bdeed3ed587a145fd9e8028642","source":{"kind":"arxiv","id":"2605.24583","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.24583","created_at":"2026-05-26T01:03:47Z"},{"alias_kind":"arxiv_version","alias_value":"2605.24583v1","created_at":"2026-05-26T01:03:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.24583","created_at":"2026-05-26T01:03:47Z"},{"alias_kind":"pith_short_12","alias_value":"YUKIMGA4UUUG","created_at":"2026-05-26T01:03:47Z"},{"alias_kind":"pith_short_16","alias_value":"YUKIMGA4UUUGPKVB","created_at":"2026-05-26T01:03:47Z"},{"alias_kind":"pith_short_8","alias_value":"YUKIMGA4","created_at":"2026-05-26T01:03:47Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:YUKIMGA4UUUGPKVBUPZ7TVOSBL","target":"record","payload":{"canonical_record":{"source":{"id":"2605.24583","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-23T13:47:17Z","cross_cats_sorted":["cs.CL","stat.ML"],"title_canon_sha256":"90e9f16fa30048d951bde5e3340d28d44f87b49a4c5a8a8d175b98a9e6a7f3cf","abstract_canon_sha256":"a641204a06f864c11f5a8dddd05d40e6ff8a7c1324d2e44d8c4e89b13f97e1b1"},"schema_version":"1.0"},"canonical_sha256":"c51486181ca52867aaa1a3f3f9d5d20ae7d938bdeed3ed587a145fd9e8028642","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-26T01:03:47.463910Z","signature_b64":"Ms5agH87fHHW2SXaHgzTV9tk8OBZhL1tmNudyeBrmCMwGwK35VyafL4Z5nIINl2dYAubAdE7zXzIkbRTTTz3CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c51486181ca52867aaa1a3f3f9d5d20ae7d938bdeed3ed587a145fd9e8028642","last_reissued_at":"2026-05-26T01:03:47.463062Z","signature_status":"signed_v1","first_computed_at":"2026-05-26T01:03:47.463062Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.24583","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":"2026-05-26T01:03:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ZbtiuOS3zp8I+WGCJRBP3M8+nc3jmaGSPthtsnDMH99w1PVKU7WsChOlOsvcfph919umAeCuPGlw3H5/Xn81BQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T00:40:50.891372Z"},"content_sha256":"b67dc5945836333deda57c88c51761f25e53b0a7ec22df8388b1503a3af6ecd1","schema_version":"1.0","event_id":"sha256:b67dc5945836333deda57c88c51761f25e53b0a7ec22df8388b1503a3af6ecd1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:YUKIMGA4UUUGPKVBUPZ7TVOSBL","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"An Effective-Rank Audit of Alignment-Induced Activation Shifts: Confound Control, Constructive Calibration, and Limits","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","stat.ML"],"primary_cat":"cs.LG","authors_text":"Yuki Nakamura","submitted_at":"2026-05-23T13:47:17Z","abstract_excerpt":"We audit alignment-induced shifts in residual-stream activations of three open-weight instruction-tuned LLMs (Llama-3.1-8B-Instruct, Gemma-2-9B-it, Qwen-2.5-7B-Instruct) using the effective rank of the alignment modification matrix on safety-relevant inputs, rho_eps := rank_eps(M_Ds)/d, which formalizes the single-refusal-direction observation of Arditi et al. (2024) as a continuous quantity. The paper has three contributions. (1) Confound-controlled measurement: a four-variant decomposition (M_naive, M_template, M_aligned, M_DiD) separates chat-template formatting, alignment-stage shift, and "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.24583","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/2605.24583/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":"2026-05-26T01:03:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"duhlGhGI2zrYgXueUf42fcLd9g4asIlxPf1ribRldPL7GLIC96+hYzrqhwg5JBCYDV97ahILOAt/3sWqOLj/CA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T00:40:50.892241Z"},"content_sha256":"87f3349732cc19ea6af94e7fe0de0e3150c10f9d2c4b4b33eea6c8fb6b03becb","schema_version":"1.0","event_id":"sha256:87f3349732cc19ea6af94e7fe0de0e3150c10f9d2c4b4b33eea6c8fb6b03becb"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YUKIMGA4UUUGPKVBUPZ7TVOSBL/bundle.json","state_url":"https://pith.science/pith/YUKIMGA4UUUGPKVBUPZ7TVOSBL/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YUKIMGA4UUUGPKVBUPZ7TVOSBL/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":"2026-05-31T00:40:50Z","links":{"resolver":"https://pith.science/pith/YUKIMGA4UUUGPKVBUPZ7TVOSBL","bundle":"https://pith.science/pith/YUKIMGA4UUUGPKVBUPZ7TVOSBL/bundle.json","state":"https://pith.science/pith/YUKIMGA4UUUGPKVBUPZ7TVOSBL/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YUKIMGA4UUUGPKVBUPZ7TVOSBL/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:YUKIMGA4UUUGPKVBUPZ7TVOSBL","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":"a641204a06f864c11f5a8dddd05d40e6ff8a7c1324d2e44d8c4e89b13f97e1b1","cross_cats_sorted":["cs.CL","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-23T13:47:17Z","title_canon_sha256":"90e9f16fa30048d951bde5e3340d28d44f87b49a4c5a8a8d175b98a9e6a7f3cf"},"schema_version":"1.0","source":{"id":"2605.24583","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.24583","created_at":"2026-05-26T01:03:47Z"},{"alias_kind":"arxiv_version","alias_value":"2605.24583v1","created_at":"2026-05-26T01:03:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.24583","created_at":"2026-05-26T01:03:47Z"},{"alias_kind":"pith_short_12","alias_value":"YUKIMGA4UUUG","created_at":"2026-05-26T01:03:47Z"},{"alias_kind":"pith_short_16","alias_value":"YUKIMGA4UUUGPKVB","created_at":"2026-05-26T01:03:47Z"},{"alias_kind":"pith_short_8","alias_value":"YUKIMGA4","created_at":"2026-05-26T01:03:47Z"}],"graph_snapshots":[{"event_id":"sha256:87f3349732cc19ea6af94e7fe0de0e3150c10f9d2c4b4b33eea6c8fb6b03becb","target":"graph","created_at":"2026-05-26T01:03:47Z","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/2605.24583/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We audit alignment-induced shifts in residual-stream activations of three open-weight instruction-tuned LLMs (Llama-3.1-8B-Instruct, Gemma-2-9B-it, Qwen-2.5-7B-Instruct) using the effective rank of the alignment modification matrix on safety-relevant inputs, rho_eps := rank_eps(M_Ds)/d, which formalizes the single-refusal-direction observation of Arditi et al. (2024) as a continuous quantity. The paper has three contributions. (1) Confound-controlled measurement: a four-variant decomposition (M_naive, M_template, M_aligned, M_DiD) separates chat-template formatting, alignment-stage shift, and ","authors_text":"Yuki Nakamura","cross_cats":["cs.CL","stat.ML"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-23T13:47:17Z","title":"An Effective-Rank Audit of Alignment-Induced Activation Shifts: Confound Control, Constructive Calibration, and Limits"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.24583","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:b67dc5945836333deda57c88c51761f25e53b0a7ec22df8388b1503a3af6ecd1","target":"record","created_at":"2026-05-26T01:03:47Z","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":"a641204a06f864c11f5a8dddd05d40e6ff8a7c1324d2e44d8c4e89b13f97e1b1","cross_cats_sorted":["cs.CL","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-23T13:47:17Z","title_canon_sha256":"90e9f16fa30048d951bde5e3340d28d44f87b49a4c5a8a8d175b98a9e6a7f3cf"},"schema_version":"1.0","source":{"id":"2605.24583","kind":"arxiv","version":1}},"canonical_sha256":"c51486181ca52867aaa1a3f3f9d5d20ae7d938bdeed3ed587a145fd9e8028642","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c51486181ca52867aaa1a3f3f9d5d20ae7d938bdeed3ed587a145fd9e8028642","first_computed_at":"2026-05-26T01:03:47.463062Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-26T01:03:47.463062Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Ms5agH87fHHW2SXaHgzTV9tk8OBZhL1tmNudyeBrmCMwGwK35VyafL4Z5nIINl2dYAubAdE7zXzIkbRTTTz3CQ==","signature_status":"signed_v1","signed_at":"2026-05-26T01:03:47.463910Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.24583","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b67dc5945836333deda57c88c51761f25e53b0a7ec22df8388b1503a3af6ecd1","sha256:87f3349732cc19ea6af94e7fe0de0e3150c10f9d2c4b4b33eea6c8fb6b03becb"],"state_sha256":"c31df2a46b99505981e3c0b7667024507985f7a280ae1b691c702e1a2fbd6a81"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"btNQgPfZq0NQMtFKiODvobFEiujlpEvBfQ5knkNJDJXBtAohWEdkhF9mz/3UhiP7ZF5S6MBiEl9T4o4hMt6JDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-31T00:40:50.897001Z","bundle_sha256":"5ceac93cedd7179a7569d16f0675182dee0e656a9a73eb27f4e457a5c53c6f34"}}