{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:624A3Y2EUX53JZP3RSK62XVVDP","short_pith_number":"pith:624A3Y2E","schema_version":"1.0","canonical_sha256":"f6b80de344a5fbb4e5fb8c95ed5eb51bdc6d129fbc0a77f627d3e5d49dbf7de6","source":{"kind":"arxiv","id":"2009.10924","version":2},"attestation_state":"computed","paper":{"title":"FusionStitching: Boosting Memory Intensive Computations for Deep Learning Workloads","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.DC","authors_text":"Feiwen Zhu, Guoping Long, Jun Yang, Kai Zhu, Lansong Diao, Pengzhan Zhao, Wei Lin, Wenyi Zhao, Zhen Zheng","submitted_at":"2020-09-23T04:00:53Z","abstract_excerpt":"We show in this work that memory intensive computations can result in severe performance problems due to off-chip memory access and CPU-GPU context switch overheads in a wide range of deep learning models. For this problem, current just-in-time (JIT) kernel fusion and code generation techniques have limitations, such as rough fusion plan exploration strategies and limited code generation ability. We propose FusionStitching, a deep learning compiler capable of fusing memory intensive operators, with varied data dependencies and non-homogeneous parallelism, into large GPU kernels to reduce globa"},"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":"2009.10924","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2020-09-23T04:00:53Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"fbbcf55390bd5baf7f79a916201e5ba0154dc2f7873182430829594339beb3ea","abstract_canon_sha256":"94600b0bfc1cb26fc8130de92b8059e6d8331ca697632ef3659511378ed8c945"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:41:37.371808Z","signature_b64":"em3b/kVNKx6AWejUhzjUwLHFTnXXrPyYLZXuh0kwxD7oyN1WhHtFe8h+hLpQ9+DllMbrQe4VjcNfJlTpTC6dCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f6b80de344a5fbb4e5fb8c95ed5eb51bdc6d129fbc0a77f627d3e5d49dbf7de6","last_reissued_at":"2026-07-05T03:41:37.371340Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:41:37.371340Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"FusionStitching: Boosting Memory Intensive Computations for Deep Learning Workloads","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.DC","authors_text":"Feiwen Zhu, Guoping Long, Jun Yang, Kai Zhu, Lansong Diao, Pengzhan Zhao, Wei Lin, Wenyi Zhao, Zhen Zheng","submitted_at":"2020-09-23T04:00:53Z","abstract_excerpt":"We show in this work that memory intensive computations can result in severe performance problems due to off-chip memory access and CPU-GPU context switch overheads in a wide range of deep learning models. For this problem, current just-in-time (JIT) kernel fusion and code generation techniques have limitations, such as rough fusion plan exploration strategies and limited code generation ability. We propose FusionStitching, a deep learning compiler capable of fusing memory intensive operators, with varied data dependencies and non-homogeneous parallelism, into large GPU kernels to reduce globa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2009.10924","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/2009.10924/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":"2009.10924","created_at":"2026-07-05T03:41:37.371405+00:00"},{"alias_kind":"arxiv_version","alias_value":"2009.10924v2","created_at":"2026-07-05T03:41:37.371405+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2009.10924","created_at":"2026-07-05T03:41:37.371405+00:00"},{"alias_kind":"pith_short_12","alias_value":"624A3Y2EUX53","created_at":"2026-07-05T03:41:37.371405+00:00"},{"alias_kind":"pith_short_16","alias_value":"624A3Y2EUX53JZP3","created_at":"2026-07-05T03:41:37.371405+00:00"},{"alias_kind":"pith_short_8","alias_value":"624A3Y2E","created_at":"2026-07-05T03:41:37.371405+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2608.03537","citing_title":"ComFuse: Fusing Complex Memory-Intensive Subgraphs with Compute-Intensive Kernels For Modern GPU Architectures","ref_index":43,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/624A3Y2EUX53JZP3RSK62XVVDP","json":"https://pith.science/pith/624A3Y2EUX53JZP3RSK62XVVDP.json","graph_json":"https://pith.science/api/pith-number/624A3Y2EUX53JZP3RSK62XVVDP/graph.json","events_json":"https://pith.science/api/pith-number/624A3Y2EUX53JZP3RSK62XVVDP/events.json","paper":"https://pith.science/paper/624A3Y2E"},"agent_actions":{"view_html":"https://pith.science/pith/624A3Y2EUX53JZP3RSK62XVVDP","download_json":"https://pith.science/pith/624A3Y2EUX53JZP3RSK62XVVDP.json","view_paper":"https://pith.science/paper/624A3Y2E","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2009.10924&json=true","fetch_graph":"https://pith.science/api/pith-number/624A3Y2EUX53JZP3RSK62XVVDP/graph.json","fetch_events":"https://pith.science/api/pith-number/624A3Y2EUX53JZP3RSK62XVVDP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/624A3Y2EUX53JZP3RSK62XVVDP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/624A3Y2EUX53JZP3RSK62XVVDP/action/storage_attestation","attest_author":"https://pith.science/pith/624A3Y2EUX53JZP3RSK62XVVDP/action/author_attestation","sign_citation":"https://pith.science/pith/624A3Y2EUX53JZP3RSK62XVVDP/action/citation_signature","submit_replication":"https://pith.science/pith/624A3Y2EUX53JZP3RSK62XVVDP/action/replication_record"}},"created_at":"2026-07-05T03:41:37.371405+00:00","updated_at":"2026-07-05T03:41:37.371405+00:00"}