{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:M7F4YA3MIPNK6N33AMXFOB6XOM","short_pith_number":"pith:M7F4YA3M","schema_version":"1.0","canonical_sha256":"67cbcc036c43daaf377b032e5707d7730a3250264d745f85742e73fc2ced0431","source":{"kind":"arxiv","id":"2004.14601","version":3},"attestation_state":"computed","paper":{"title":"Learning Music Helps You Read: Using Transfer to Study Linguistic Structure in Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dan Jurafsky, Isabel Papadimitriou","submitted_at":"2020-04-30T06:24:03Z","abstract_excerpt":"We propose transfer learning as a method for analyzing the encoding of grammatical structure in neural language models. We train LSTMs on non-linguistic data and evaluate their performance on natural language to assess which kinds of data induce generalizable structural features that LSTMs can use for natural language. We find that training on non-linguistic data with latent structure (MIDI music or Java code) improves test performance on natural language, despite no overlap in surface form or vocabulary. To pinpoint the kinds of abstract structure that models may be encoding to lead to this i"},"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":"2004.14601","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-04-30T06:24:03Z","cross_cats_sorted":[],"title_canon_sha256":"d79884ea7d69daa731044e59d0486d3f2add0d01eda2aad746e4ebec0175bfbe","abstract_canon_sha256":"e48a8e4e51286e2b73b8f49a917d7ae0f0ba823ad62f0172119732beb91a82b9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:47:43.787226Z","signature_b64":"zziwE8SO+nY8l+97RfUe16/vsvyUSmFCr29ysBWlVTVkmjaktUCiLxkOVELtQEn/zP+SIrKsH70EAU85YbrxCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"67cbcc036c43daaf377b032e5707d7730a3250264d745f85742e73fc2ced0431","last_reissued_at":"2026-07-05T01:47:43.786830Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:47:43.786830Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning Music Helps You Read: Using Transfer to Study Linguistic Structure in Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dan Jurafsky, Isabel Papadimitriou","submitted_at":"2020-04-30T06:24:03Z","abstract_excerpt":"We propose transfer learning as a method for analyzing the encoding of grammatical structure in neural language models. We train LSTMs on non-linguistic data and evaluate their performance on natural language to assess which kinds of data induce generalizable structural features that LSTMs can use for natural language. We find that training on non-linguistic data with latent structure (MIDI music or Java code) improves test performance on natural language, despite no overlap in surface form or vocabulary. To pinpoint the kinds of abstract structure that models may be encoding to lead to this i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2004.14601","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/2004.14601/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":"2004.14601","created_at":"2026-07-05T01:47:43.786881+00:00"},{"alias_kind":"arxiv_version","alias_value":"2004.14601v3","created_at":"2026-07-05T01:47:43.786881+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2004.14601","created_at":"2026-07-05T01:47:43.786881+00:00"},{"alias_kind":"pith_short_12","alias_value":"M7F4YA3MIPNK","created_at":"2026-07-05T01:47:43.786881+00:00"},{"alias_kind":"pith_short_16","alias_value":"M7F4YA3MIPNK6N33","created_at":"2026-07-05T01:47:43.786881+00:00"},{"alias_kind":"pith_short_8","alias_value":"M7F4YA3M","created_at":"2026-07-05T01:47:43.786881+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.02293","citing_title":"AI as a Tool for Simulation-Based Experiments in Literary Studies","ref_index":54,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/M7F4YA3MIPNK6N33AMXFOB6XOM","json":"https://pith.science/pith/M7F4YA3MIPNK6N33AMXFOB6XOM.json","graph_json":"https://pith.science/api/pith-number/M7F4YA3MIPNK6N33AMXFOB6XOM/graph.json","events_json":"https://pith.science/api/pith-number/M7F4YA3MIPNK6N33AMXFOB6XOM/events.json","paper":"https://pith.science/paper/M7F4YA3M"},"agent_actions":{"view_html":"https://pith.science/pith/M7F4YA3MIPNK6N33AMXFOB6XOM","download_json":"https://pith.science/pith/M7F4YA3MIPNK6N33AMXFOB6XOM.json","view_paper":"https://pith.science/paper/M7F4YA3M","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2004.14601&json=true","fetch_graph":"https://pith.science/api/pith-number/M7F4YA3MIPNK6N33AMXFOB6XOM/graph.json","fetch_events":"https://pith.science/api/pith-number/M7F4YA3MIPNK6N33AMXFOB6XOM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/M7F4YA3MIPNK6N33AMXFOB6XOM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/M7F4YA3MIPNK6N33AMXFOB6XOM/action/storage_attestation","attest_author":"https://pith.science/pith/M7F4YA3MIPNK6N33AMXFOB6XOM/action/author_attestation","sign_citation":"https://pith.science/pith/M7F4YA3MIPNK6N33AMXFOB6XOM/action/citation_signature","submit_replication":"https://pith.science/pith/M7F4YA3MIPNK6N33AMXFOB6XOM/action/replication_record"}},"created_at":"2026-07-05T01:47:43.786881+00:00","updated_at":"2026-07-05T01:47:43.786881+00:00"}