{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:STSXVC2YMURURUQVXHSO32WXNO","short_pith_number":"pith:STSXVC2Y","schema_version":"1.0","canonical_sha256":"94e57a8b58652348d215b9e4edead76b9d90aad9c5be6273953ff6ffe4e13cba","source":{"kind":"arxiv","id":"2110.14883","version":3},"attestation_state":"computed","paper":{"title":"Colossal-AI: A Unified Deep Learning System For Large-Scale Parallel Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.CV","cs.DC"],"primary_cat":"cs.LG","authors_text":"Boxiang Wang, Haichen Huang, Hongxin Liu, Jiarui Fang, Shenggui Li, Yang You, Yuliang Liu, Zhengda Bian","submitted_at":"2021-10-28T04:45:55Z","abstract_excerpt":"The success of Transformer models has pushed the deep learning model scale to billions of parameters. Due to the limited memory resource of a single GPU, However, the best practice for choosing the optimal parallel strategy is still lacking, since it requires domain expertise in both deep learning and parallel computing.\n  The Colossal-AI system addressed the above challenge by introducing a unified interface to scale your sequential code of model training to distributed environments. It supports parallel training methods such as data, pipeline, tensor, and sequence parallelism, as well as het"},"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":"2110.14883","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-10-28T04:45:55Z","cross_cats_sorted":["cs.AI","cs.CL","cs.CV","cs.DC"],"title_canon_sha256":"9e19fff060786e2b0715d67db4bcfe17b565dd7b9bba9904294ad4ac622add38","abstract_canon_sha256":"6569b4e55c1c4b1d08da5ad71a3a088c3a326f93691ba86c8f482e8de92905dc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:57:18.411060Z","signature_b64":"WQTfozaENLQZkhmigYjPlfz+Hm0SfMbMfeAIJQWRWnjcxQ9Har453NtsmTIDDNJt3/fYtb7dFBwzqd5Ck7/RCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"94e57a8b58652348d215b9e4edead76b9d90aad9c5be6273953ff6ffe4e13cba","last_reissued_at":"2026-07-05T06:57:18.410575Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:57:18.410575Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Colossal-AI: A Unified Deep Learning System For Large-Scale Parallel Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.CV","cs.DC"],"primary_cat":"cs.LG","authors_text":"Boxiang Wang, Haichen Huang, Hongxin Liu, Jiarui Fang, Shenggui Li, Yang You, Yuliang Liu, Zhengda Bian","submitted_at":"2021-10-28T04:45:55Z","abstract_excerpt":"The success of Transformer models has pushed the deep learning model scale to billions of parameters. Due to the limited memory resource of a single GPU, However, the best practice for choosing the optimal parallel strategy is still lacking, since it requires domain expertise in both deep learning and parallel computing.\n  The Colossal-AI system addressed the above challenge by introducing a unified interface to scale your sequential code of model training to distributed environments. It supports parallel training methods such as data, pipeline, tensor, and sequence parallelism, as well as het"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.14883","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/2110.14883/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":"2110.14883","created_at":"2026-07-05T06:57:18.410630+00:00"},{"alias_kind":"arxiv_version","alias_value":"2110.14883v3","created_at":"2026-07-05T06:57:18.410630+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.14883","created_at":"2026-07-05T06:57:18.410630+00:00"},{"alias_kind":"pith_short_12","alias_value":"STSXVC2YMURU","created_at":"2026-07-05T06:57:18.410630+00:00"},{"alias_kind":"pith_short_16","alias_value":"STSXVC2YMURURUQV","created_at":"2026-07-05T06:57:18.410630+00:00"},{"alias_kind":"pith_short_8","alias_value":"STSXVC2Y","created_at":"2026-07-05T06:57:18.410630+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2307.06435","citing_title":"A Comprehensive Overview of Large Language Models","ref_index":84,"is_internal_anchor":false},{"citing_arxiv_id":"2305.16264","citing_title":"Scaling Data-Constrained Language Models","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2303.18223","citing_title":"A Survey of Large Language Models","ref_index":207,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/STSXVC2YMURURUQVXHSO32WXNO","json":"https://pith.science/pith/STSXVC2YMURURUQVXHSO32WXNO.json","graph_json":"https://pith.science/api/pith-number/STSXVC2YMURURUQVXHSO32WXNO/graph.json","events_json":"https://pith.science/api/pith-number/STSXVC2YMURURUQVXHSO32WXNO/events.json","paper":"https://pith.science/paper/STSXVC2Y"},"agent_actions":{"view_html":"https://pith.science/pith/STSXVC2YMURURUQVXHSO32WXNO","download_json":"https://pith.science/pith/STSXVC2YMURURUQVXHSO32WXNO.json","view_paper":"https://pith.science/paper/STSXVC2Y","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2110.14883&json=true","fetch_graph":"https://pith.science/api/pith-number/STSXVC2YMURURUQVXHSO32WXNO/graph.json","fetch_events":"https://pith.science/api/pith-number/STSXVC2YMURURUQVXHSO32WXNO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/STSXVC2YMURURUQVXHSO32WXNO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/STSXVC2YMURURUQVXHSO32WXNO/action/storage_attestation","attest_author":"https://pith.science/pith/STSXVC2YMURURUQVXHSO32WXNO/action/author_attestation","sign_citation":"https://pith.science/pith/STSXVC2YMURURUQVXHSO32WXNO/action/citation_signature","submit_replication":"https://pith.science/pith/STSXVC2YMURURUQVXHSO32WXNO/action/replication_record"}},"created_at":"2026-07-05T06:57:18.410630+00:00","updated_at":"2026-07-05T06:57:18.410630+00:00"}