{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:DA2C5CZTQWHPZQJOTL4TCZS6G6","short_pith_number":"pith:DA2C5CZT","schema_version":"1.0","canonical_sha256":"18342e8b33858efcc12e9af931665e379526af67f6ad270deac00452776bf2d1","source":{"kind":"arxiv","id":"2111.05897","version":3},"attestation_state":"computed","paper":{"title":"Persia: An Open, Hybrid System Scaling Deep Learning-based Recommenders up to 100 Trillion Parameters","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.LG","authors_text":"Binhang Yuan, Ce Zhang, Chengchun Shu, Chengjun Liu, Congfei Zhang, Dongying Kong, Hai Yu, Haodong Lyu, Haonan Li, Honghuan Wu, Jianying Lin, Ji Liu, Jingru Xie, Lei Chen, Lei Sun, Lei Yuan, Mingnan Luo, Renjie Huang, Sen Yang, Xiangru Lian, Xing Dong, Xuefeng Zhu, Xuezhong Qiu, Yiqiao Liao, Yongjun He, Yulong Wang, Zhishan Liu","submitted_at":"2021-11-10T19:40:25Z","abstract_excerpt":"Deep learning based models have dominated the current landscape of production recommender systems. Furthermore, recent years have witnessed an exponential growth of the model scale--from Google's 2016 model with 1 billion parameters to the latest Facebook's model with 12 trillion parameters. Significant quality boost has come with each jump of the model capacity, which makes us believe the era of 100 trillion parameters is around the corner. However, the training of such models is challenging even within industrial scale data centers. This difficulty is inherited from the staggering heterogene"},"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":"2111.05897","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.LG","submitted_at":"2021-11-10T19:40:25Z","cross_cats_sorted":["cs.DC"],"title_canon_sha256":"aa3541c8340a6bbdf299c28d528e2b8df0ba4a8e17a5b8396eb5318b461b9d5d","abstract_canon_sha256":"ce0f57bbc040596327b8380691eb110d6765e4ddb1ea5636d5582a7c242302ee"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:34:08.646341Z","signature_b64":"Q323WBgfqno5wP3spJ1f0eCkq6KP9koJ0durfc4JBeiv3sJjP3gXgmhWahVr/teSEFHLU0rttfqC60eS+qzUBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"18342e8b33858efcc12e9af931665e379526af67f6ad270deac00452776bf2d1","last_reissued_at":"2026-07-05T03:34:08.645854Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:34:08.645854Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Persia: An Open, Hybrid System Scaling Deep Learning-based Recommenders up to 100 Trillion Parameters","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.LG","authors_text":"Binhang Yuan, Ce Zhang, Chengchun Shu, Chengjun Liu, Congfei Zhang, Dongying Kong, Hai Yu, Haodong Lyu, Haonan Li, Honghuan Wu, Jianying Lin, Ji Liu, Jingru Xie, Lei Chen, Lei Sun, Lei Yuan, Mingnan Luo, Renjie Huang, Sen Yang, Xiangru Lian, Xing Dong, Xuefeng Zhu, Xuezhong Qiu, Yiqiao Liao, Yongjun He, Yulong Wang, Zhishan Liu","submitted_at":"2021-11-10T19:40:25Z","abstract_excerpt":"Deep learning based models have dominated the current landscape of production recommender systems. Furthermore, recent years have witnessed an exponential growth of the model scale--from Google's 2016 model with 1 billion parameters to the latest Facebook's model with 12 trillion parameters. Significant quality boost has come with each jump of the model capacity, which makes us believe the era of 100 trillion parameters is around the corner. However, the training of such models is challenging even within industrial scale data centers. This difficulty is inherited from the staggering heterogene"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2111.05897","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/2111.05897/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":"2111.05897","created_at":"2026-07-05T03:34:08.645911+00:00"},{"alias_kind":"arxiv_version","alias_value":"2111.05897v3","created_at":"2026-07-05T03:34:08.645911+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2111.05897","created_at":"2026-07-05T03:34:08.645911+00:00"},{"alias_kind":"pith_short_12","alias_value":"DA2C5CZTQWHP","created_at":"2026-07-05T03:34:08.645911+00:00"},{"alias_kind":"pith_short_16","alias_value":"DA2C5CZTQWHPZQJO","created_at":"2026-07-05T03:34:08.645911+00:00"},{"alias_kind":"pith_short_8","alias_value":"DA2C5CZT","created_at":"2026-07-05T03:34:08.645911+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/DA2C5CZTQWHPZQJOTL4TCZS6G6","json":"https://pith.science/pith/DA2C5CZTQWHPZQJOTL4TCZS6G6.json","graph_json":"https://pith.science/api/pith-number/DA2C5CZTQWHPZQJOTL4TCZS6G6/graph.json","events_json":"https://pith.science/api/pith-number/DA2C5CZTQWHPZQJOTL4TCZS6G6/events.json","paper":"https://pith.science/paper/DA2C5CZT"},"agent_actions":{"view_html":"https://pith.science/pith/DA2C5CZTQWHPZQJOTL4TCZS6G6","download_json":"https://pith.science/pith/DA2C5CZTQWHPZQJOTL4TCZS6G6.json","view_paper":"https://pith.science/paper/DA2C5CZT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2111.05897&json=true","fetch_graph":"https://pith.science/api/pith-number/DA2C5CZTQWHPZQJOTL4TCZS6G6/graph.json","fetch_events":"https://pith.science/api/pith-number/DA2C5CZTQWHPZQJOTL4TCZS6G6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DA2C5CZTQWHPZQJOTL4TCZS6G6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DA2C5CZTQWHPZQJOTL4TCZS6G6/action/storage_attestation","attest_author":"https://pith.science/pith/DA2C5CZTQWHPZQJOTL4TCZS6G6/action/author_attestation","sign_citation":"https://pith.science/pith/DA2C5CZTQWHPZQJOTL4TCZS6G6/action/citation_signature","submit_replication":"https://pith.science/pith/DA2C5CZTQWHPZQJOTL4TCZS6G6/action/replication_record"}},"created_at":"2026-07-05T03:34:08.645911+00:00","updated_at":"2026-07-05T03:34:08.645911+00:00"}