{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:WVCPJ7DFE2VY4HPZU2G4CROND4","short_pith_number":"pith:WVCPJ7DF","schema_version":"1.0","canonical_sha256":"b544f4fc6526ab8e1df9a68dc145cd1f2b01c9cf294eae838dc1cda66159d5a7","source":{"kind":"arxiv","id":"2605.00333","version":2},"attestation_state":"computed","paper":{"title":"Borrowed Geometry: Cross-Distribution Head-Importance Fingerprints of Frozen Pretrained Gemma 4 31B","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Frozen text-pretrained transformer weights transfer to robotic and memory tasks through a thin trainable interface without modifying the core model.","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Abay Bektursun","submitted_at":"2026-05-01T01:23:37Z","abstract_excerpt":"Frozen Gemma 4 31B weights pretrained exclusively on text, unmodified, transfer through a thin trainable interface to non-text modalities the substrate has never processed. On the L24--L29 slice (192 attention heads), an English-text TxtCopy attention probe (95 sentences) and per-head ablation impact on four non-language token-pattern tasks (binary copy, associative recall, 1D cellular automaton Rule 90, binary addition) jointly classify four heads -- L26.28, L27.28, L27.2, L27.3 -- as top-tier on both signals. The slice-level joint coincidence is significant under hypergeometric null ($P = 0."},"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":"2605.00333","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-01T01:23:37Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"b90d0fe3051a66c4bdbdf8f7076876895ec97d5ccca464eb5c6fec71f83d6de2","abstract_canon_sha256":"c42832faa1011e543464576bba8cdbfe5f0e9aaea9274cbf7c22b3b5121c35df"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T01:05:14.868311Z","signature_b64":"QX2nUK6KjuXo4iRXFh1aDBSwSWyqnLiGcS5La0JV0Kw6D7n4J4HADh5FH1TEKimZ5xhjQ/pnGNEh3Lgp2m6cAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b544f4fc6526ab8e1df9a68dc145cd1f2b01c9cf294eae838dc1cda66159d5a7","last_reissued_at":"2026-05-20T01:05:14.867261Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T01:05:14.867261Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Borrowed Geometry: Cross-Distribution Head-Importance Fingerprints of Frozen Pretrained Gemma 4 31B","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Frozen text-pretrained transformer weights transfer to robotic and memory tasks through a thin trainable interface without modifying the core model.","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Abay Bektursun","submitted_at":"2026-05-01T01:23:37Z","abstract_excerpt":"Frozen Gemma 4 31B weights pretrained exclusively on text, unmodified, transfer through a thin trainable interface to non-text modalities the substrate has never processed. On the L24--L29 slice (192 attention heads), an English-text TxtCopy attention probe (95 sentences) and per-head ablation impact on four non-language token-pattern tasks (binary copy, associative recall, 1D cellular automaton Rule 90, binary addition) jointly classify four heads -- L26.28, L27.28, L27.2, L27.3 -- as top-tier on both signals. The slice-level joint coincidence is significant under hypergeometric null ($P = 0."},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Frozen Gemma 4 31B weights pretrained exclusively on text tokens, unmodified, transfer across modality boundaries through a thin trainable interface.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the geometry captured by text-only pretraining remains useful and transferable to non-text modalities without any modification to the core frozen weights.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Frozen text-pretrained transformer weights transfer across modalities through a thin interface, achieving SOTA on a robotic task and parity on decision-making with far fewer trainable parameters.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Frozen text-pretrained transformer weights transfer to robotic and memory tasks through a thin trainable interface without modifying the core model.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"207760f4c8b4cafd86db2d9ab3f7eecedd68784e6906469ffbc03096cd85f416"},"source":{"id":"2605.00333","kind":"arxiv","version":2},"verdict":{"id":"05f7c2ff-cec2-49f0-a49c-b5617c649e24","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-09T20:27:37.081861Z","strongest_claim":"Frozen Gemma 4 31B weights pretrained exclusively on text tokens, unmodified, transfer across modality boundaries through a thin trainable interface.","one_line_summary":"Frozen text-pretrained transformer weights transfer across modalities through a thin interface, achieving SOTA on a robotic task and parity on decision-making with far fewer trainable parameters.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the geometry captured by text-only pretraining remains useful and transferable to non-text modalities without any modification to the core frozen weights.","pith_extraction_headline":"Frozen text-pretrained transformer weights transfer to robotic and memory tasks through a thin trainable interface without modifying the core model."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.00333/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_compliance","ran_at":"2026-05-19T18:15:56.124031Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"567a48c8ba3438466c5ca81a4a531495591bb5fbeddc58ec7ef5dfb71b0a80a6"},"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":"2605.00333","created_at":"2026-05-20T01:05:14.867737+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.00333v2","created_at":"2026-05-20T01:05:14.867737+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.00333","created_at":"2026-05-20T01:05:14.867737+00:00"},{"alias_kind":"pith_short_12","alias_value":"WVCPJ7DFE2VY","created_at":"2026-05-20T01:05:14.867737+00:00"},{"alias_kind":"pith_short_16","alias_value":"WVCPJ7DFE2VY4HPZ","created_at":"2026-05-20T01:05:14.867737+00:00"},{"alias_kind":"pith_short_8","alias_value":"WVCPJ7DF","created_at":"2026-05-20T01:05:14.867737+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/WVCPJ7DFE2VY4HPZU2G4CROND4","json":"https://pith.science/pith/WVCPJ7DFE2VY4HPZU2G4CROND4.json","graph_json":"https://pith.science/api/pith-number/WVCPJ7DFE2VY4HPZU2G4CROND4/graph.json","events_json":"https://pith.science/api/pith-number/WVCPJ7DFE2VY4HPZU2G4CROND4/events.json","paper":"https://pith.science/paper/WVCPJ7DF"},"agent_actions":{"view_html":"https://pith.science/pith/WVCPJ7DFE2VY4HPZU2G4CROND4","download_json":"https://pith.science/pith/WVCPJ7DFE2VY4HPZU2G4CROND4.json","view_paper":"https://pith.science/paper/WVCPJ7DF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.00333&json=true","fetch_graph":"https://pith.science/api/pith-number/WVCPJ7DFE2VY4HPZU2G4CROND4/graph.json","fetch_events":"https://pith.science/api/pith-number/WVCPJ7DFE2VY4HPZU2G4CROND4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WVCPJ7DFE2VY4HPZU2G4CROND4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WVCPJ7DFE2VY4HPZU2G4CROND4/action/storage_attestation","attest_author":"https://pith.science/pith/WVCPJ7DFE2VY4HPZU2G4CROND4/action/author_attestation","sign_citation":"https://pith.science/pith/WVCPJ7DFE2VY4HPZU2G4CROND4/action/citation_signature","submit_replication":"https://pith.science/pith/WVCPJ7DFE2VY4HPZU2G4CROND4/action/replication_record"}},"created_at":"2026-05-20T01:05:14.867737+00:00","updated_at":"2026-05-20T01:05:14.867737+00:00"}