{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:AC74XNPKGR37PKHSZ6KTFFPOIV","short_pith_number":"pith:AC74XNPK","schema_version":"1.0","canonical_sha256":"00bfcbb5ea3477f7a8f2cf953295ee45525f7131cfd436df908dbd93a957f6fb","source":{"kind":"arxiv","id":"2101.01332","version":2},"attestation_state":"computed","paper":{"title":"Equality Saturation for Tensor Graph Superoptimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.AI","authors_text":"Jacques Pienaar, Max Willsey, Phitchaya Mangpo Phothilimtha, Sudip Roy, Yichen Yang, Yisu Remy Wang","submitted_at":"2021-01-05T03:30:35Z","abstract_excerpt":"One of the major optimizations employed in deep learning frameworks is graph rewriting. Production frameworks rely on heuristics to decide if rewrite rules should be applied and in which order. Prior research has shown that one can discover more optimal tensor computation graphs if we search for a better sequence of substitutions instead of relying on heuristics. However, we observe that existing approaches for tensor graph superoptimization both in production and research frameworks apply substitutions in a sequential manner. Such sequential search methods are sensitive to the order in which "},"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":"2101.01332","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2021-01-05T03:30:35Z","cross_cats_sorted":["cs.DC"],"title_canon_sha256":"195e714303c3b4c308c4ce64019583201944e147bb1e483fa2d7f850374eb8be","abstract_canon_sha256":"c00067af894d479c0c0466b13aa0943e394c9e72691d43f822731386abe28c00"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:24:00.456492Z","signature_b64":"TWGpj+pBUtMrU1b2MieOmrWbCO8mb/2nLx9Aqmp5bURgcb4jZcBd8ZmqSjXQdxAOjQTpl+3AXQcG+CsVa1YuBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"00bfcbb5ea3477f7a8f2cf953295ee45525f7131cfd436df908dbd93a957f6fb","last_reissued_at":"2026-07-05T02:24:00.456044Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:24:00.456044Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Equality Saturation for Tensor Graph Superoptimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.AI","authors_text":"Jacques Pienaar, Max Willsey, Phitchaya Mangpo Phothilimtha, Sudip Roy, Yichen Yang, Yisu Remy Wang","submitted_at":"2021-01-05T03:30:35Z","abstract_excerpt":"One of the major optimizations employed in deep learning frameworks is graph rewriting. Production frameworks rely on heuristics to decide if rewrite rules should be applied and in which order. Prior research has shown that one can discover more optimal tensor computation graphs if we search for a better sequence of substitutions instead of relying on heuristics. However, we observe that existing approaches for tensor graph superoptimization both in production and research frameworks apply substitutions in a sequential manner. Such sequential search methods are sensitive to the order in which "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2101.01332","kind":"arxiv","version":2},"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/2101.01332/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":"2101.01332","created_at":"2026-07-05T02:24:00.456093+00:00"},{"alias_kind":"arxiv_version","alias_value":"2101.01332v2","created_at":"2026-07-05T02:24:00.456093+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2101.01332","created_at":"2026-07-05T02:24:00.456093+00:00"},{"alias_kind":"pith_short_12","alias_value":"AC74XNPKGR37","created_at":"2026-07-05T02:24:00.456093+00:00"},{"alias_kind":"pith_short_16","alias_value":"AC74XNPKGR37PKHS","created_at":"2026-07-05T02:24:00.456093+00:00"},{"alias_kind":"pith_short_8","alias_value":"AC74XNPK","created_at":"2026-07-05T02:24:00.456093+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.06747","citing_title":"Tensor Algebraic Property Skeletons: Amplifying Property-Based Testing for AI Compilers","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2605.19005","citing_title":"Rewrite System Showdown: Stochastic Search vs. EqSat","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2605.19005","citing_title":"Rewrite System Showdown: Stochastic Search vs. EqSat","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2511.15000","citing_title":"Bonsai: Compiling Queries to Pruned Tree Traversals","ref_index":88,"is_internal_anchor":false},{"citing_arxiv_id":"2511.20782","citing_title":"Optimism in Equality Saturation","ref_index":53,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14825","citing_title":"Nautilus: An Auto-Scheduling Tensor Compiler for Efficient Tiled GPU Kernels","ref_index":46,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17364","citing_title":"LLM-Guided Strategy Synthesis for Scalable Equality Saturation","ref_index":42,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AC74XNPKGR37PKHSZ6KTFFPOIV","json":"https://pith.science/pith/AC74XNPKGR37PKHSZ6KTFFPOIV.json","graph_json":"https://pith.science/api/pith-number/AC74XNPKGR37PKHSZ6KTFFPOIV/graph.json","events_json":"https://pith.science/api/pith-number/AC74XNPKGR37PKHSZ6KTFFPOIV/events.json","paper":"https://pith.science/paper/AC74XNPK"},"agent_actions":{"view_html":"https://pith.science/pith/AC74XNPKGR37PKHSZ6KTFFPOIV","download_json":"https://pith.science/pith/AC74XNPKGR37PKHSZ6KTFFPOIV.json","view_paper":"https://pith.science/paper/AC74XNPK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2101.01332&json=true","fetch_graph":"https://pith.science/api/pith-number/AC74XNPKGR37PKHSZ6KTFFPOIV/graph.json","fetch_events":"https://pith.science/api/pith-number/AC74XNPKGR37PKHSZ6KTFFPOIV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AC74XNPKGR37PKHSZ6KTFFPOIV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AC74XNPKGR37PKHSZ6KTFFPOIV/action/storage_attestation","attest_author":"https://pith.science/pith/AC74XNPKGR37PKHSZ6KTFFPOIV/action/author_attestation","sign_citation":"https://pith.science/pith/AC74XNPKGR37PKHSZ6KTFFPOIV/action/citation_signature","submit_replication":"https://pith.science/pith/AC74XNPKGR37PKHSZ6KTFFPOIV/action/replication_record"}},"created_at":"2026-07-05T02:24:00.456093+00:00","updated_at":"2026-07-05T02:24:00.456093+00:00"}