{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:SRDUIZLPY3Q5QKQYTECKCOH5RR","short_pith_number":"pith:SRDUIZLP","schema_version":"1.0","canonical_sha256":"944744656fc6e1d82a189904a138fd8c72abae0f692d973ab81e5598b06d4ad0","source":{"kind":"arxiv","id":"2607.21356","version":1},"attestation_state":"computed","paper":{"title":"Emergent Misalignment Recruits a Pre-existing Persona Subspace","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Mohammed Suhail B Nadaf","submitted_at":"2026-07-23T14:19:28Z","abstract_excerpt":"Fine-tuning an aligned language model on a narrow stream of bad advice can make it broadly misaligned on questions unrelated to the training data, a phenomenon called emergent misalignment. We ask why the narrow lesson generalizes at all, and we find that narrow fine-tuning recruits a persona structure that is present in the model before the fine-tune exists. From a frozen instruction-tuned model (Qwen2.5-14B-Instruct) we extract per-domain persona subspaces by contrastive teacher forcing and find that 4 unrelated domains share one low-rank core at 657x a random-subspace null, with 82% of that"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2607.21356","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-23T14:19:28Z","cross_cats_sorted":[],"title_canon_sha256":"1799a1189953bbc79daf27e9328878f605fa88b32e25a24042618a19cfd9e4a4","abstract_canon_sha256":"29d13a1aa2eda3ce862a5ba12ad93a96bea8193b2351528ea6db58dfdf8ebb82"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-24T01:24:27.353433Z","signature_b64":"j2On6J3IVDFwqW98B5GjWKv89QlnbCnw0xTqRa+Q2bVAwatCOrn/uQsyeGdZHN3Ofy2zkm8MLTFHrpNoKHxCDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"944744656fc6e1d82a189904a138fd8c72abae0f692d973ab81e5598b06d4ad0","last_reissued_at":"2026-07-24T01:24:27.352538Z","signature_status":"signed_v1","first_computed_at":"2026-07-24T01:24:27.352538Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Emergent Misalignment Recruits a Pre-existing Persona Subspace","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Mohammed Suhail B Nadaf","submitted_at":"2026-07-23T14:19:28Z","abstract_excerpt":"Fine-tuning an aligned language model on a narrow stream of bad advice can make it broadly misaligned on questions unrelated to the training data, a phenomenon called emergent misalignment. We ask why the narrow lesson generalizes at all, and we find that narrow fine-tuning recruits a persona structure that is present in the model before the fine-tune exists. From a frozen instruction-tuned model (Qwen2.5-14B-Instruct) we extract per-domain persona subspaces by contrastive teacher forcing and find that 4 unrelated domains share one low-rank core at 657x a random-subspace null, with 82% of that"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.21356","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/2607.21356/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2607.21356","created_at":"2026-07-24T01:24:27.352997+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.21356v1","created_at":"2026-07-24T01:24:27.352997+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.21356","created_at":"2026-07-24T01:24:27.352997+00:00"},{"alias_kind":"pith_short_12","alias_value":"SRDUIZLPY3Q5","created_at":"2026-07-24T01:24:27.352997+00:00"},{"alias_kind":"pith_short_16","alias_value":"SRDUIZLPY3Q5QKQY","created_at":"2026-07-24T01:24:27.352997+00:00"},{"alias_kind":"pith_short_8","alias_value":"SRDUIZLP","created_at":"2026-07-24T01:24:27.352997+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SRDUIZLPY3Q5QKQYTECKCOH5RR","json":"https://pith.science/pith/SRDUIZLPY3Q5QKQYTECKCOH5RR.json","graph_json":"https://pith.science/api/pith-number/SRDUIZLPY3Q5QKQYTECKCOH5RR/graph.json","events_json":"https://pith.science/api/pith-number/SRDUIZLPY3Q5QKQYTECKCOH5RR/events.json","paper":"https://pith.science/paper/SRDUIZLP"},"agent_actions":{"view_html":"https://pith.science/pith/SRDUIZLPY3Q5QKQYTECKCOH5RR","download_json":"https://pith.science/pith/SRDUIZLPY3Q5QKQYTECKCOH5RR.json","view_paper":"https://pith.science/paper/SRDUIZLP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.21356&json=true","fetch_graph":"https://pith.science/api/pith-number/SRDUIZLPY3Q5QKQYTECKCOH5RR/graph.json","fetch_events":"https://pith.science/api/pith-number/SRDUIZLPY3Q5QKQYTECKCOH5RR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SRDUIZLPY3Q5QKQYTECKCOH5RR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SRDUIZLPY3Q5QKQYTECKCOH5RR/action/storage_attestation","attest_author":"https://pith.science/pith/SRDUIZLPY3Q5QKQYTECKCOH5RR/action/author_attestation","sign_citation":"https://pith.science/pith/SRDUIZLPY3Q5QKQYTECKCOH5RR/action/citation_signature","submit_replication":"https://pith.science/pith/SRDUIZLPY3Q5QKQYTECKCOH5RR/action/replication_record"}},"created_at":"2026-07-24T01:24:27.352997+00:00","updated_at":"2026-07-24T01:24:27.352997+00:00"}