{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:N43I6K46LHMS4FZ72ACEMZNGUJ","short_pith_number":"pith:N43I6K46","schema_version":"1.0","canonical_sha256":"6f368f2b9e59d92e173fd0044665a6a253abf094c5afb0fb7da9a184583dae23","source":{"kind":"arxiv","id":"2309.10706","version":2},"attestation_state":"computed","paper":{"title":"OpenBA: An Open-sourced 15B Bilingual Asymmetric seq2seq Model Pre-trained from Scratch","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dan Qiao, Guodong Zhou, Guohong Fu, Juntao Li, Min Zhang, Pei Guo, Pinzheng Wang, Qiaoming Zhu, Wangjie You, Wenliang Chen, Yuyang Ding, Zecheng Tang","submitted_at":"2023-09-19T15:46:40Z","abstract_excerpt":"Large language models (LLMs) with billions of parameters have demonstrated outstanding performance on various natural language processing tasks. This report presents OpenBA, an open-sourced 15B bilingual asymmetric seq2seq model, to contribute an LLM variant to the Chinese-oriented open-source model community. We enhance OpenBA with effective and efficient techniques as well as adopt a three-stage training strategy to train the model from scratch. Our solution can also achieve very competitive performance with only 380B tokens, which is better than LLaMA-70B on the BELEBELE benchmark, BLOOM-17"},"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":"2309.10706","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-09-19T15:46:40Z","cross_cats_sorted":[],"title_canon_sha256":"a6e7358497fe822df972a667807985309b4a02442ee2dd85ad1dc7e2dbe8d41a","abstract_canon_sha256":"77e08cffdb0ea2d2d3379a04ad8197a0ba4913f137ec0ed491ae4ff5ca2b716b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:39:19.526910Z","signature_b64":"XFxahL6b/iHf2B9qlR8ES5lemRzuTtxEOIGJHOcYzpRubCZoFNuzdltTZfUmHRwTxFdXlcIak4eNyVTyo+9BAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6f368f2b9e59d92e173fd0044665a6a253abf094c5afb0fb7da9a184583dae23","last_reissued_at":"2026-07-05T09:39:19.526402Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:39:19.526402Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"OpenBA: An Open-sourced 15B Bilingual Asymmetric seq2seq Model Pre-trained from Scratch","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dan Qiao, Guodong Zhou, Guohong Fu, Juntao Li, Min Zhang, Pei Guo, Pinzheng Wang, Qiaoming Zhu, Wangjie You, Wenliang Chen, Yuyang Ding, Zecheng Tang","submitted_at":"2023-09-19T15:46:40Z","abstract_excerpt":"Large language models (LLMs) with billions of parameters have demonstrated outstanding performance on various natural language processing tasks. This report presents OpenBA, an open-sourced 15B bilingual asymmetric seq2seq model, to contribute an LLM variant to the Chinese-oriented open-source model community. We enhance OpenBA with effective and efficient techniques as well as adopt a three-stage training strategy to train the model from scratch. Our solution can also achieve very competitive performance with only 380B tokens, which is better than LLaMA-70B on the BELEBELE benchmark, BLOOM-17"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.10706","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/2309.10706/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":"2309.10706","created_at":"2026-07-05T09:39:19.526463+00:00"},{"alias_kind":"arxiv_version","alias_value":"2309.10706v2","created_at":"2026-07-05T09:39:19.526463+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.10706","created_at":"2026-07-05T09:39:19.526463+00:00"},{"alias_kind":"pith_short_12","alias_value":"N43I6K46LHMS","created_at":"2026-07-05T09:39:19.526463+00:00"},{"alias_kind":"pith_short_16","alias_value":"N43I6K46LHMS4FZ7","created_at":"2026-07-05T09:39:19.526463+00:00"},{"alias_kind":"pith_short_8","alias_value":"N43I6K46","created_at":"2026-07-05T09:39:19.526463+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/N43I6K46LHMS4FZ72ACEMZNGUJ","json":"https://pith.science/pith/N43I6K46LHMS4FZ72ACEMZNGUJ.json","graph_json":"https://pith.science/api/pith-number/N43I6K46LHMS4FZ72ACEMZNGUJ/graph.json","events_json":"https://pith.science/api/pith-number/N43I6K46LHMS4FZ72ACEMZNGUJ/events.json","paper":"https://pith.science/paper/N43I6K46"},"agent_actions":{"view_html":"https://pith.science/pith/N43I6K46LHMS4FZ72ACEMZNGUJ","download_json":"https://pith.science/pith/N43I6K46LHMS4FZ72ACEMZNGUJ.json","view_paper":"https://pith.science/paper/N43I6K46","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2309.10706&json=true","fetch_graph":"https://pith.science/api/pith-number/N43I6K46LHMS4FZ72ACEMZNGUJ/graph.json","fetch_events":"https://pith.science/api/pith-number/N43I6K46LHMS4FZ72ACEMZNGUJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/N43I6K46LHMS4FZ72ACEMZNGUJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/N43I6K46LHMS4FZ72ACEMZNGUJ/action/storage_attestation","attest_author":"https://pith.science/pith/N43I6K46LHMS4FZ72ACEMZNGUJ/action/author_attestation","sign_citation":"https://pith.science/pith/N43I6K46LHMS4FZ72ACEMZNGUJ/action/citation_signature","submit_replication":"https://pith.science/pith/N43I6K46LHMS4FZ72ACEMZNGUJ/action/replication_record"}},"created_at":"2026-07-05T09:39:19.526463+00:00","updated_at":"2026-07-05T09:39:19.526463+00:00"}