{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:OL2F6CB6X4ZWJW5FH2OEMNALMZ","short_pith_number":"pith:OL2F6CB6","schema_version":"1.0","canonical_sha256":"72f45f083ebf3364dba53e9c46340b664c25d804b5186e50371d2757e359ad34","source":{"kind":"arxiv","id":"2607.02945","version":1},"attestation_state":"computed","paper":{"title":"Optimus: A Generic Operator-Level PyTorch Model Transformation Framework","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.PF","authors_text":"Brian Hirsh, Chi-Keung Luk, Daohang Shi, Edward Yang, Elias Ellison, Huaqing Xiong, Jason Ansel, Jia Chen Ren, Jia Liu, Jiaqi Xu, Junqing Zhou, Menglu Yu, Mingming Ding, Oguz Ulgen, Quanyu Zhu, Ruilin Chen, Shuai Yang, Will Feng, Xu Zhao, Yanbo Liang, Yuhang Yang, Yuzhen Huang","submitted_at":"2026-07-03T04:40:04Z","abstract_excerpt":"In large-scale industrial applications, deep learning models that power recommendation and ranking have complex and diverse model architectures. These models are continuously developed and refined by large teams of machine learning engineers, rendering manual optimization infeasible. Consequently, graph-based optimization techniques have become an industry standard for boosting performance, with PyTorch FX transformations leading the charge. These transformations typically rely on a set of human-engineered module-level rewrite rules which are not scalable to diverse model architectures. To add"},"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":"2607.02945","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.PF","submitted_at":"2026-07-03T04:40:04Z","cross_cats_sorted":[],"title_canon_sha256":"c10fa54b5d0a428c59717168c26cce75600b64a32ad3416fbb0b81b084750a7b","abstract_canon_sha256":"52683aeaa0b7f9fbc350374b8aefa1c123ddce56ac4f3139a96c72b8f59beca3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-07T01:16:37.550434Z","signature_b64":"9BHUw7wDAnTREI5Lch5sHt7VQKb6ty5R+pO0UrD8WFY3xHy8J4Bi0R5bVhdho+Pp70y8U+JhT8RIbEdop/qkCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"72f45f083ebf3364dba53e9c46340b664c25d804b5186e50371d2757e359ad34","last_reissued_at":"2026-07-07T01:16:37.549943Z","signature_status":"signed_v1","first_computed_at":"2026-07-07T01:16:37.549943Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Optimus: A Generic Operator-Level PyTorch Model Transformation Framework","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.PF","authors_text":"Brian Hirsh, Chi-Keung Luk, Daohang Shi, Edward Yang, Elias Ellison, Huaqing Xiong, Jason Ansel, Jia Chen Ren, Jia Liu, Jiaqi Xu, Junqing Zhou, Menglu Yu, Mingming Ding, Oguz Ulgen, Quanyu Zhu, Ruilin Chen, Shuai Yang, Will Feng, Xu Zhao, Yanbo Liang, Yuhang Yang, Yuzhen Huang","submitted_at":"2026-07-03T04:40:04Z","abstract_excerpt":"In large-scale industrial applications, deep learning models that power recommendation and ranking have complex and diverse model architectures. These models are continuously developed and refined by large teams of machine learning engineers, rendering manual optimization infeasible. Consequently, graph-based optimization techniques have become an industry standard for boosting performance, with PyTorch FX transformations leading the charge. These transformations typically rely on a set of human-engineered module-level rewrite rules which are not scalable to diverse model architectures. To add"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.02945","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/2607.02945/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":"2607.02945","created_at":"2026-07-07T01:16:37.550001+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.02945v1","created_at":"2026-07-07T01:16:37.550001+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.02945","created_at":"2026-07-07T01:16:37.550001+00:00"},{"alias_kind":"pith_short_12","alias_value":"OL2F6CB6X4ZW","created_at":"2026-07-07T01:16:37.550001+00:00"},{"alias_kind":"pith_short_16","alias_value":"OL2F6CB6X4ZWJW5F","created_at":"2026-07-07T01:16:37.550001+00:00"},{"alias_kind":"pith_short_8","alias_value":"OL2F6CB6","created_at":"2026-07-07T01:16:37.550001+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/OL2F6CB6X4ZWJW5FH2OEMNALMZ","json":"https://pith.science/pith/OL2F6CB6X4ZWJW5FH2OEMNALMZ.json","graph_json":"https://pith.science/api/pith-number/OL2F6CB6X4ZWJW5FH2OEMNALMZ/graph.json","events_json":"https://pith.science/api/pith-number/OL2F6CB6X4ZWJW5FH2OEMNALMZ/events.json","paper":"https://pith.science/paper/OL2F6CB6"},"agent_actions":{"view_html":"https://pith.science/pith/OL2F6CB6X4ZWJW5FH2OEMNALMZ","download_json":"https://pith.science/pith/OL2F6CB6X4ZWJW5FH2OEMNALMZ.json","view_paper":"https://pith.science/paper/OL2F6CB6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.02945&json=true","fetch_graph":"https://pith.science/api/pith-number/OL2F6CB6X4ZWJW5FH2OEMNALMZ/graph.json","fetch_events":"https://pith.science/api/pith-number/OL2F6CB6X4ZWJW5FH2OEMNALMZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OL2F6CB6X4ZWJW5FH2OEMNALMZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OL2F6CB6X4ZWJW5FH2OEMNALMZ/action/storage_attestation","attest_author":"https://pith.science/pith/OL2F6CB6X4ZWJW5FH2OEMNALMZ/action/author_attestation","sign_citation":"https://pith.science/pith/OL2F6CB6X4ZWJW5FH2OEMNALMZ/action/citation_signature","submit_replication":"https://pith.science/pith/OL2F6CB6X4ZWJW5FH2OEMNALMZ/action/replication_record"}},"created_at":"2026-07-07T01:16:37.550001+00:00","updated_at":"2026-07-07T01:16:37.550001+00:00"}