{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:27UNDFL35K54OKRAWQFLM62RCS","short_pith_number":"pith:27UNDFL3","schema_version":"1.0","canonical_sha256":"d7e8d1957beabbc72a20b40ab67b5114be02422f1979f0939fb457876a1a1246","source":{"kind":"arxiv","id":"2411.17607","version":2},"attestation_state":"computed","paper":{"title":"Scaling Speech-Text Pre-training with Synthetic Interleaved Data","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Aohan Zeng, Jie Tang, Lei Zhang, Mingdao Liu, Shengmin Jiang, Yuxiao Dong, Zhengxiao Du","submitted_at":"2024-11-26T17:19:09Z","abstract_excerpt":"Speech language models (SpeechLMs) accept speech input and produce speech output, allowing for more natural human-computer interaction compared to text-based large language models (LLMs). Traditional approaches for developing SpeechLMs are constrained by the limited availability of unsupervised speech data and parallel speech-text data, which are significantly less abundant than text pre-training data, thereby limiting their scalability as LLMs. We propose a novel approach to scaling speech-text pre-training by leveraging large-scale synthetic interleaved data derived from text corpora, elimin"},"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":"2411.17607","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-11-26T17:19:09Z","cross_cats_sorted":["cs.SD","eess.AS"],"title_canon_sha256":"4252cc0f0f2fe568e6fddf88038e729d9e0e2501a4583802c8b1a744fc02b809","abstract_canon_sha256":"0bfb85baffb69941b38118eb88506b84d196f34b9a468a590109ec0874b99c52"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:43:12.020295Z","signature_b64":"E8GtW6Qw6c044lUfH8zFRpys4I4yWoSSo8w/eHIuZC5pQhRw7PR+J+qXsPoxdGp7TnCuZUtu3lvuvx9O5A2ICg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d7e8d1957beabbc72a20b40ab67b5114be02422f1979f0939fb457876a1a1246","last_reissued_at":"2026-07-05T09:43:12.019749Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:43:12.019749Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Scaling Speech-Text Pre-training with Synthetic Interleaved Data","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Aohan Zeng, Jie Tang, Lei Zhang, Mingdao Liu, Shengmin Jiang, Yuxiao Dong, Zhengxiao Du","submitted_at":"2024-11-26T17:19:09Z","abstract_excerpt":"Speech language models (SpeechLMs) accept speech input and produce speech output, allowing for more natural human-computer interaction compared to text-based large language models (LLMs). Traditional approaches for developing SpeechLMs are constrained by the limited availability of unsupervised speech data and parallel speech-text data, which are significantly less abundant than text pre-training data, thereby limiting their scalability as LLMs. We propose a novel approach to scaling speech-text pre-training by leveraging large-scale synthetic interleaved data derived from text corpora, elimin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.17607","kind":"arxiv","version":2},"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/2411.17607/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":"2411.17607","created_at":"2026-07-05T09:43:12.019836+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.17607v2","created_at":"2026-07-05T09:43:12.019836+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.17607","created_at":"2026-07-05T09:43:12.019836+00:00"},{"alias_kind":"pith_short_12","alias_value":"27UNDFL35K54","created_at":"2026-07-05T09:43:12.019836+00:00"},{"alias_kind":"pith_short_16","alias_value":"27UNDFL35K54OKRA","created_at":"2026-07-05T09:43:12.019836+00:00"},{"alias_kind":"pith_short_8","alias_value":"27UNDFL3","created_at":"2026-07-05T09:43:12.019836+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2509.03526","citing_title":"Enhancing Speech Large Language Models through Reinforced Behavior Alignment","ref_index":57,"is_internal_anchor":false},{"citing_arxiv_id":"2412.02612","citing_title":"GLM-4-Voice: Towards Intelligent and Human-Like End-to-End Spoken Chatbot","ref_index":46,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06765","citing_title":"VITA-QinYu: Expressive Spoken Language Model for Role-Playing and Singing","ref_index":50,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/27UNDFL35K54OKRAWQFLM62RCS","json":"https://pith.science/pith/27UNDFL35K54OKRAWQFLM62RCS.json","graph_json":"https://pith.science/api/pith-number/27UNDFL35K54OKRAWQFLM62RCS/graph.json","events_json":"https://pith.science/api/pith-number/27UNDFL35K54OKRAWQFLM62RCS/events.json","paper":"https://pith.science/paper/27UNDFL3"},"agent_actions":{"view_html":"https://pith.science/pith/27UNDFL35K54OKRAWQFLM62RCS","download_json":"https://pith.science/pith/27UNDFL35K54OKRAWQFLM62RCS.json","view_paper":"https://pith.science/paper/27UNDFL3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.17607&json=true","fetch_graph":"https://pith.science/api/pith-number/27UNDFL35K54OKRAWQFLM62RCS/graph.json","fetch_events":"https://pith.science/api/pith-number/27UNDFL35K54OKRAWQFLM62RCS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/27UNDFL35K54OKRAWQFLM62RCS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/27UNDFL35K54OKRAWQFLM62RCS/action/storage_attestation","attest_author":"https://pith.science/pith/27UNDFL35K54OKRAWQFLM62RCS/action/author_attestation","sign_citation":"https://pith.science/pith/27UNDFL35K54OKRAWQFLM62RCS/action/citation_signature","submit_replication":"https://pith.science/pith/27UNDFL35K54OKRAWQFLM62RCS/action/replication_record"}},"created_at":"2026-07-05T09:43:12.019836+00:00","updated_at":"2026-07-05T09:43:12.019836+00:00"}