{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:ZUKUR5F366K4OUOGTWS22ARQUG","short_pith_number":"pith:ZUKUR5F3","schema_version":"1.0","canonical_sha256":"cd1548f4bbf795c751c69da5ad0230a1a8425a26bbf29b972bbda768af74ea64","source":{"kind":"arxiv","id":"2502.15592","version":1},"attestation_state":"computed","paper":{"title":"Generalizing From Short to Long: Effective Data Synthesis for Long-Context Instruction Tuning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Alexandra Birch, Fei Yuan, Hanxu Hu, Jiajun Chen, Pinzhen Chen, Shujian Huang, Wenhao Zhu","submitted_at":"2025-02-21T17:02:40Z","abstract_excerpt":"Long-context modelling for large language models (LLMs) has been a key area of recent research because many real world use cases require reasoning over longer inputs such as documents. The focus of research into modelling long context has been on how to model position and there has been little investigation into other important aspects of language modelling such as instruction tuning. Long context training examples are challenging and expensive to create and use. In this paper, we investigate how to design instruction data for the post-training phase of a long context pre-trained model: how mu"},"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":"2502.15592","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-21T17:02:40Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"daf77c14383bafe01d4b0016b0a27d4dcc3175be3986ca42293eeae1f20c22ca","abstract_canon_sha256":"530908de9c52b46e6e3dd39f444be6a6c314faf504b0d78e6b16e5a69740eabc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:18:04.010518Z","signature_b64":"Osziw9xQaTwIDj6OEBZetNC2Kp1w1OauLSSP44WFxzw9Rwgamwd4Sk5wZ4XGqkxEfpr4VM6D2qqheA9jQkTzAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cd1548f4bbf795c751c69da5ad0230a1a8425a26bbf29b972bbda768af74ea64","last_reissued_at":"2026-07-05T10:18:04.010007Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:18:04.010007Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Generalizing From Short to Long: Effective Data Synthesis for Long-Context Instruction Tuning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Alexandra Birch, Fei Yuan, Hanxu Hu, Jiajun Chen, Pinzhen Chen, Shujian Huang, Wenhao Zhu","submitted_at":"2025-02-21T17:02:40Z","abstract_excerpt":"Long-context modelling for large language models (LLMs) has been a key area of recent research because many real world use cases require reasoning over longer inputs such as documents. The focus of research into modelling long context has been on how to model position and there has been little investigation into other important aspects of language modelling such as instruction tuning. Long context training examples are challenging and expensive to create and use. In this paper, we investigate how to design instruction data for the post-training phase of a long context pre-trained model: how mu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.15592","kind":"arxiv","version":1},"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/2502.15592/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":"2502.15592","created_at":"2026-07-05T10:18:04.010069+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.15592v1","created_at":"2026-07-05T10:18:04.010069+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.15592","created_at":"2026-07-05T10:18:04.010069+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZUKUR5F366K4","created_at":"2026-07-05T10:18:04.010069+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZUKUR5F366K4OUOG","created_at":"2026-07-05T10:18:04.010069+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZUKUR5F3","created_at":"2026-07-05T10:18:04.010069+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.02176","citing_title":"Adam's Law: Textual Frequency Law on Large Language Models","ref_index":40,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZUKUR5F366K4OUOGTWS22ARQUG","json":"https://pith.science/pith/ZUKUR5F366K4OUOGTWS22ARQUG.json","graph_json":"https://pith.science/api/pith-number/ZUKUR5F366K4OUOGTWS22ARQUG/graph.json","events_json":"https://pith.science/api/pith-number/ZUKUR5F366K4OUOGTWS22ARQUG/events.json","paper":"https://pith.science/paper/ZUKUR5F3"},"agent_actions":{"view_html":"https://pith.science/pith/ZUKUR5F366K4OUOGTWS22ARQUG","download_json":"https://pith.science/pith/ZUKUR5F366K4OUOGTWS22ARQUG.json","view_paper":"https://pith.science/paper/ZUKUR5F3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.15592&json=true","fetch_graph":"https://pith.science/api/pith-number/ZUKUR5F366K4OUOGTWS22ARQUG/graph.json","fetch_events":"https://pith.science/api/pith-number/ZUKUR5F366K4OUOGTWS22ARQUG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZUKUR5F366K4OUOGTWS22ARQUG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZUKUR5F366K4OUOGTWS22ARQUG/action/storage_attestation","attest_author":"https://pith.science/pith/ZUKUR5F366K4OUOGTWS22ARQUG/action/author_attestation","sign_citation":"https://pith.science/pith/ZUKUR5F366K4OUOGTWS22ARQUG/action/citation_signature","submit_replication":"https://pith.science/pith/ZUKUR5F366K4OUOGTWS22ARQUG/action/replication_record"}},"created_at":"2026-07-05T10:18:04.010069+00:00","updated_at":"2026-07-05T10:18:04.010069+00:00"}