{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:KZXYJTXJMSRFY2IT4UBCHGP27H","short_pith_number":"pith:KZXYJTXJ","schema_version":"1.0","canonical_sha256":"566f84cee964a25c6913e5022399faf9d980cb07897b61353b1ff27ab6d23be9","source":{"kind":"arxiv","id":"2601.00065","version":3},"attestation_state":"computed","paper":{"title":"When the Same Coefficients Reach Different Places: Asymmetric Realizability in Transplanting Tokenizers across Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.CR"],"primary_cat":"cs.LG","authors_text":"Jing Gao, Matt Fredrikson, Weichen Yu, Xiaoqian Wang, Xiaoze Liu","submitted_at":"2025-12-31T19:00:03Z","abstract_excerpt":"Tokenizer transplant in cross-vocabulary model composition reconstructs donor-only embedding rows as weighted combinations over shared lexical anchors and reuses those coefficients on the base. We identify a structural geometric property of this reconstruction: the same coefficient vector reaches different sets in the donor and base anchor spans, an \\emph{asymmetric realizability} gap. Across 65 donor-base pairs under OMP, with cross-operator validation on CLP, WECHSEL, and FOCUS, we construct \\textit{breaker tokens}: single coefficient vectors that remain statistically inert in the donor anch"},"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":"2601.00065","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-12-31T19:00:03Z","cross_cats_sorted":["cs.CL","cs.CR"],"title_canon_sha256":"7b9a64714d80034dbb4db62d7d708020890bb34ae888fa8e44a72dd5d306b7a3","abstract_canon_sha256":"3869311d9a9762a35d1525843783982738a0e523aabe73d453c346b76fe36904"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T01:05:00.793035Z","signature_b64":"rxnV10q0hKjkW/GBC+P30ZZoNoBtEoLi0+UsfehkFrl/8/sxlpRLok3inR5CoNi9SxdE4xee7QdqYALmRhh5AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"566f84cee964a25c6913e5022399faf9d980cb07897b61353b1ff27ab6d23be9","last_reissued_at":"2026-05-29T01:05:00.792117Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T01:05:00.792117Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"When the Same Coefficients Reach Different Places: Asymmetric Realizability in Transplanting Tokenizers across Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.CR"],"primary_cat":"cs.LG","authors_text":"Jing Gao, Matt Fredrikson, Weichen Yu, Xiaoqian Wang, Xiaoze Liu","submitted_at":"2025-12-31T19:00:03Z","abstract_excerpt":"Tokenizer transplant in cross-vocabulary model composition reconstructs donor-only embedding rows as weighted combinations over shared lexical anchors and reuses those coefficients on the base. We identify a structural geometric property of this reconstruction: the same coefficient vector reaches different sets in the donor and base anchor spans, an \\emph{asymmetric realizability} gap. Across 65 donor-base pairs under OMP, with cross-operator validation on CLP, WECHSEL, and FOCUS, we construct \\textit{breaker tokens}: single coefficient vectors that remain statistically inert in the donor anch"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2601.00065","kind":"arxiv","version":3},"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/2601.00065/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":"2601.00065","created_at":"2026-05-29T01:05:00.792232+00:00"},{"alias_kind":"arxiv_version","alias_value":"2601.00065v3","created_at":"2026-05-29T01:05:00.792232+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.00065","created_at":"2026-05-29T01:05:00.792232+00:00"},{"alias_kind":"pith_short_12","alias_value":"KZXYJTXJMSRF","created_at":"2026-05-29T01:05:00.792232+00:00"},{"alias_kind":"pith_short_16","alias_value":"KZXYJTXJMSRFY2IT","created_at":"2026-05-29T01:05:00.792232+00:00"},{"alias_kind":"pith_short_8","alias_value":"KZXYJTXJ","created_at":"2026-05-29T01:05:00.792232+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/KZXYJTXJMSRFY2IT4UBCHGP27H","json":"https://pith.science/pith/KZXYJTXJMSRFY2IT4UBCHGP27H.json","graph_json":"https://pith.science/api/pith-number/KZXYJTXJMSRFY2IT4UBCHGP27H/graph.json","events_json":"https://pith.science/api/pith-number/KZXYJTXJMSRFY2IT4UBCHGP27H/events.json","paper":"https://pith.science/paper/KZXYJTXJ"},"agent_actions":{"view_html":"https://pith.science/pith/KZXYJTXJMSRFY2IT4UBCHGP27H","download_json":"https://pith.science/pith/KZXYJTXJMSRFY2IT4UBCHGP27H.json","view_paper":"https://pith.science/paper/KZXYJTXJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2601.00065&json=true","fetch_graph":"https://pith.science/api/pith-number/KZXYJTXJMSRFY2IT4UBCHGP27H/graph.json","fetch_events":"https://pith.science/api/pith-number/KZXYJTXJMSRFY2IT4UBCHGP27H/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KZXYJTXJMSRFY2IT4UBCHGP27H/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KZXYJTXJMSRFY2IT4UBCHGP27H/action/storage_attestation","attest_author":"https://pith.science/pith/KZXYJTXJMSRFY2IT4UBCHGP27H/action/author_attestation","sign_citation":"https://pith.science/pith/KZXYJTXJMSRFY2IT4UBCHGP27H/action/citation_signature","submit_replication":"https://pith.science/pith/KZXYJTXJMSRFY2IT4UBCHGP27H/action/replication_record"}},"created_at":"2026-05-29T01:05:00.792232+00:00","updated_at":"2026-05-29T01:05:00.792232+00:00"}