{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:OGLKDY74HHYTS32D5AQV64OQES","short_pith_number":"pith:OGLKDY74","schema_version":"1.0","canonical_sha256":"7196a1e3fc39f1396f43e8215f71d0248d3cab995ceb6bf76d934917491a6481","source":{"kind":"arxiv","id":"2201.12023","version":3},"attestation_state":"computed","paper":{"title":"Alpa: Automating Inter- and Intra-Operator Parallelism for Distributed Deep Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC","cs.PL"],"primary_cat":"cs.LG","authors_text":"Danyang Zhuo, Eric P. Xing, Hao Zhang, Ion Stoica, Joseph E. Gonzalez, Lianmin Zheng, Yanping Huang, Yida Wang, Yonghao Zhuang, Yuanzhong Xu, Zhifeng Chen, Zhuohan Li","submitted_at":"2022-01-28T10:13:35Z","abstract_excerpt":"Alpa automates model-parallel training of large deep learning (DL) models by generating execution plans that unify data, operator, and pipeline parallelism. Existing model-parallel training systems either require users to manually create a parallelization plan or automatically generate one from a limited space of model parallelism configurations. They do not suffice to scale out complex DL models on distributed compute devices. Alpa distributes the training of large DL models by viewing parallelisms as two hierarchical levels: inter-operator and intra-operator parallelisms. Based on it, Alpa c"},"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":"2201.12023","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-01-28T10:13:35Z","cross_cats_sorted":["cs.DC","cs.PL"],"title_canon_sha256":"359f9e37248737dc31fbf00b81cebdf2704e38d345ba691ddc5d70497fceea2d","abstract_canon_sha256":"ede133d7a5de39e50602021557959de090aba7a39b561032b93d888a31eb5157"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:35:58.456748Z","signature_b64":"4KuQzLfxpU4vy96yhaUPtDXTwHTxpQYzCcjllM/tOCTCUUwkRErDUACyzC7qvVNgLCrQAcTUBkvt3WTorIfkAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7196a1e3fc39f1396f43e8215f71d0248d3cab995ceb6bf76d934917491a6481","last_reissued_at":"2026-07-05T04:35:58.456239Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:35:58.456239Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Alpa: Automating Inter- and Intra-Operator Parallelism for Distributed Deep Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC","cs.PL"],"primary_cat":"cs.LG","authors_text":"Danyang Zhuo, Eric P. Xing, Hao Zhang, Ion Stoica, Joseph E. Gonzalez, Lianmin Zheng, Yanping Huang, Yida Wang, Yonghao Zhuang, Yuanzhong Xu, Zhifeng Chen, Zhuohan Li","submitted_at":"2022-01-28T10:13:35Z","abstract_excerpt":"Alpa automates model-parallel training of large deep learning (DL) models by generating execution plans that unify data, operator, and pipeline parallelism. Existing model-parallel training systems either require users to manually create a parallelization plan or automatically generate one from a limited space of model parallelism configurations. They do not suffice to scale out complex DL models on distributed compute devices. Alpa distributes the training of large DL models by viewing parallelisms as two hierarchical levels: inter-operator and intra-operator parallelisms. Based on it, Alpa c"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2201.12023","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/2201.12023/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":"2201.12023","created_at":"2026-07-05T04:35:58.456297+00:00"},{"alias_kind":"arxiv_version","alias_value":"2201.12023v3","created_at":"2026-07-05T04:35:58.456297+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2201.12023","created_at":"2026-07-05T04:35:58.456297+00:00"},{"alias_kind":"pith_short_12","alias_value":"OGLKDY74HHYT","created_at":"2026-07-05T04:35:58.456297+00:00"},{"alias_kind":"pith_short_16","alias_value":"OGLKDY74HHYTS32D","created_at":"2026-07-05T04:35:58.456297+00:00"},{"alias_kind":"pith_short_8","alias_value":"OGLKDY74","created_at":"2026-07-05T04:35:58.456297+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.07046","citing_title":"Voltron: Enabling Elastic Multi-Device Execution of LLM Inference for Empowered Edge Intelligence","ref_index":73,"is_internal_anchor":true},{"citing_arxiv_id":"2305.02301","citing_title":"Distilling Step-by-Step! Outperforming Larger Language Models with Less Training Data and Smaller Model Sizes","ref_index":115,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18750","citing_title":"A Readiness-Driven Runtime for Pipeline-Parallel Training under Runtime Variability","ref_index":60,"is_internal_anchor":false},{"citing_arxiv_id":"2604.23467","citing_title":"Hybrid JIT-CUDA Graph Optimization for Low-Latency Large Language Model Inference","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2210.17323","citing_title":"GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers","ref_index":17,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OGLKDY74HHYTS32D5AQV64OQES","json":"https://pith.science/pith/OGLKDY74HHYTS32D5AQV64OQES.json","graph_json":"https://pith.science/api/pith-number/OGLKDY74HHYTS32D5AQV64OQES/graph.json","events_json":"https://pith.science/api/pith-number/OGLKDY74HHYTS32D5AQV64OQES/events.json","paper":"https://pith.science/paper/OGLKDY74"},"agent_actions":{"view_html":"https://pith.science/pith/OGLKDY74HHYTS32D5AQV64OQES","download_json":"https://pith.science/pith/OGLKDY74HHYTS32D5AQV64OQES.json","view_paper":"https://pith.science/paper/OGLKDY74","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2201.12023&json=true","fetch_graph":"https://pith.science/api/pith-number/OGLKDY74HHYTS32D5AQV64OQES/graph.json","fetch_events":"https://pith.science/api/pith-number/OGLKDY74HHYTS32D5AQV64OQES/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OGLKDY74HHYTS32D5AQV64OQES/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OGLKDY74HHYTS32D5AQV64OQES/action/storage_attestation","attest_author":"https://pith.science/pith/OGLKDY74HHYTS32D5AQV64OQES/action/author_attestation","sign_citation":"https://pith.science/pith/OGLKDY74HHYTS32D5AQV64OQES/action/citation_signature","submit_replication":"https://pith.science/pith/OGLKDY74HHYTS32D5AQV64OQES/action/replication_record"}},"created_at":"2026-07-05T04:35:58.456297+00:00","updated_at":"2026-07-05T04:35:58.456297+00:00"}