{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:VWA7FD7ARCWQZSK3W4T7PKZSPJ","short_pith_number":"pith:VWA7FD7A","schema_version":"1.0","canonical_sha256":"ad81f28fe088ad0cc95bb727f7ab327a57a72ebb961d99bdfd1fb7fb2ff01e16","source":{"kind":"arxiv","id":"2406.00605","version":1},"attestation_state":"computed","paper":{"title":"LongSkywork: A Training Recipe for Efficiently Extending Context Length in Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bo Zhu, Cheng cheng, Chenxia Li, Han Fang, Liang Zeng, Liang Zhao, Lijie Wang, Liu Yang, Peng Cheng, Rui Hu, Shuicheng Yan, Tianwen Wei, Xuejie Wu, Yahui Zhou, Yimeng Gan","submitted_at":"2024-06-02T03:34:41Z","abstract_excerpt":"We introduce LongSkywork, a long-context Large Language Model (LLM) capable of processing up to 200,000 tokens. We provide a training recipe for efficiently extending context length of LLMs. We identify that the critical element in enhancing long-context processing capability is to incorporate a long-context SFT stage following the standard SFT stage. A mere 200 iterations can convert the standard SFT model into a long-context model. To reduce the effort in collecting and annotating data for long-context language modeling, we develop two novel methods for creating synthetic data. These methods"},"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":"2406.00605","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-02T03:34:41Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"1b9583a4f6300c22101ce9f68ee3e34d89ea7ccb53356595e56251dc2688194a","abstract_canon_sha256":"59432d7c41cb558b3f6b6622d10e3b554b11c694ec93f7313123f0a2e3c894e8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:26:27.995926Z","signature_b64":"i+XBjonCo3uoOhtwvZaMf32AG//IKeJjouWg5RVH+mySdcnrLkQNps8GzthmIzP5WPvuk3W30SQpSyS34v2pCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ad81f28fe088ad0cc95bb727f7ab327a57a72ebb961d99bdfd1fb7fb2ff01e16","last_reissued_at":"2026-07-05T08:26:27.995414Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:26:27.995414Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LongSkywork: A Training Recipe for Efficiently Extending Context Length in Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bo Zhu, Cheng cheng, Chenxia Li, Han Fang, Liang Zeng, Liang Zhao, Lijie Wang, Liu Yang, Peng Cheng, Rui Hu, Shuicheng Yan, Tianwen Wei, Xuejie Wu, Yahui Zhou, Yimeng Gan","submitted_at":"2024-06-02T03:34:41Z","abstract_excerpt":"We introduce LongSkywork, a long-context Large Language Model (LLM) capable of processing up to 200,000 tokens. We provide a training recipe for efficiently extending context length of LLMs. We identify that the critical element in enhancing long-context processing capability is to incorporate a long-context SFT stage following the standard SFT stage. A mere 200 iterations can convert the standard SFT model into a long-context model. To reduce the effort in collecting and annotating data for long-context language modeling, we develop two novel methods for creating synthetic data. These methods"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.00605","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/2406.00605/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":"2406.00605","created_at":"2026-07-05T08:26:27.995467+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.00605v1","created_at":"2026-07-05T08:26:27.995467+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.00605","created_at":"2026-07-05T08:26:27.995467+00:00"},{"alias_kind":"pith_short_12","alias_value":"VWA7FD7ARCWQ","created_at":"2026-07-05T08:26:27.995467+00:00"},{"alias_kind":"pith_short_16","alias_value":"VWA7FD7ARCWQZSK3","created_at":"2026-07-05T08:26:27.995467+00:00"},{"alias_kind":"pith_short_8","alias_value":"VWA7FD7A","created_at":"2026-07-05T08:26:27.995467+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.23687","citing_title":"Randomized YaRN Improves Length Generalization for Long-Context Reasoning","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2409.04777","citing_title":"Optimization Hyper-parameter Laws for Large Language Models","ref_index":40,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VWA7FD7ARCWQZSK3W4T7PKZSPJ","json":"https://pith.science/pith/VWA7FD7ARCWQZSK3W4T7PKZSPJ.json","graph_json":"https://pith.science/api/pith-number/VWA7FD7ARCWQZSK3W4T7PKZSPJ/graph.json","events_json":"https://pith.science/api/pith-number/VWA7FD7ARCWQZSK3W4T7PKZSPJ/events.json","paper":"https://pith.science/paper/VWA7FD7A"},"agent_actions":{"view_html":"https://pith.science/pith/VWA7FD7ARCWQZSK3W4T7PKZSPJ","download_json":"https://pith.science/pith/VWA7FD7ARCWQZSK3W4T7PKZSPJ.json","view_paper":"https://pith.science/paper/VWA7FD7A","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.00605&json=true","fetch_graph":"https://pith.science/api/pith-number/VWA7FD7ARCWQZSK3W4T7PKZSPJ/graph.json","fetch_events":"https://pith.science/api/pith-number/VWA7FD7ARCWQZSK3W4T7PKZSPJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VWA7FD7ARCWQZSK3W4T7PKZSPJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VWA7FD7ARCWQZSK3W4T7PKZSPJ/action/storage_attestation","attest_author":"https://pith.science/pith/VWA7FD7ARCWQZSK3W4T7PKZSPJ/action/author_attestation","sign_citation":"https://pith.science/pith/VWA7FD7ARCWQZSK3W4T7PKZSPJ/action/citation_signature","submit_replication":"https://pith.science/pith/VWA7FD7ARCWQZSK3W4T7PKZSPJ/action/replication_record"}},"created_at":"2026-07-05T08:26:27.995467+00:00","updated_at":"2026-07-05T08:26:27.995467+00:00"}