{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:5S2T7NLDWNND7OO47TP4DECN4R","short_pith_number":"pith:5S2T7NLD","schema_version":"1.0","canonical_sha256":"ecb53fb563b35a3fb9dcfcdfc1904de4557a3c69f282b08fa93d4134d6bc3b6e","source":{"kind":"arxiv","id":"2007.13828","version":2},"attestation_state":"computed","paper":{"title":"GRIP: A Graph Neural Network Accelerator Architecture","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AR","authors_text":"Christopher Re, Kevin Kiningham, Philip Levis","submitted_at":"2020-07-27T19:44:51Z","abstract_excerpt":"We present GRIP, a graph neural network accelerator architecture designed for low-latency inference. AcceleratingGNNs is challenging because they combine two distinct types of computation: arithmetic-intensive vertex-centric operations and memory-intensive edge-centric operations. GRIP splits GNN inference into a fixed set of edge- and vertex-centric execution phases that can be implemented in hardware. We then specialize each unit for the unique computational structure found in each phase.For vertex-centric phases, GRIP uses a high performance matrix multiply engine coupled with a dedicated m"},"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":"2007.13828","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AR","submitted_at":"2020-07-27T19:44:51Z","cross_cats_sorted":[],"title_canon_sha256":"0f0e7ffaa9ec25731aa36f5c17ec3cbb47312f1b7e59ad7e7a17f691e71bacae","abstract_canon_sha256":"624406a87ab0d4bf161ebc90ccde6dce72f50330c581c80a1c751aebd11fabda"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:23:26.110515Z","signature_b64":"IpMJ6Z4+O3LI+Eon5Jrr7ePaNfwbcOZLOgn2dfCTkrQvvdHG5/JYYQ7g3Yqb2zYhkJeFwoGSGCjjr+bNg3NoAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ecb53fb563b35a3fb9dcfcdfc1904de4557a3c69f282b08fa93d4134d6bc3b6e","last_reissued_at":"2026-07-05T01:23:26.110097Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:23:26.110097Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"GRIP: A Graph Neural Network Accelerator Architecture","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AR","authors_text":"Christopher Re, Kevin Kiningham, Philip Levis","submitted_at":"2020-07-27T19:44:51Z","abstract_excerpt":"We present GRIP, a graph neural network accelerator architecture designed for low-latency inference. AcceleratingGNNs is challenging because they combine two distinct types of computation: arithmetic-intensive vertex-centric operations and memory-intensive edge-centric operations. GRIP splits GNN inference into a fixed set of edge- and vertex-centric execution phases that can be implemented in hardware. We then specialize each unit for the unique computational structure found in each phase.For vertex-centric phases, GRIP uses a high performance matrix multiply engine coupled with a dedicated m"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2007.13828","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/2007.13828/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":"2007.13828","created_at":"2026-07-05T01:23:26.110161+00:00"},{"alias_kind":"arxiv_version","alias_value":"2007.13828v2","created_at":"2026-07-05T01:23:26.110161+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2007.13828","created_at":"2026-07-05T01:23:26.110161+00:00"},{"alias_kind":"pith_short_12","alias_value":"5S2T7NLDWNND","created_at":"2026-07-05T01:23:26.110161+00:00"},{"alias_kind":"pith_short_16","alias_value":"5S2T7NLDWNND7OO4","created_at":"2026-07-05T01:23:26.110161+00:00"},{"alias_kind":"pith_short_8","alias_value":"5S2T7NLD","created_at":"2026-07-05T01:23:26.110161+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.15348","citing_title":"ReInc: Scaling Training of Dynamic Graph Neural Networks","ref_index":35,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5S2T7NLDWNND7OO47TP4DECN4R","json":"https://pith.science/pith/5S2T7NLDWNND7OO47TP4DECN4R.json","graph_json":"https://pith.science/api/pith-number/5S2T7NLDWNND7OO47TP4DECN4R/graph.json","events_json":"https://pith.science/api/pith-number/5S2T7NLDWNND7OO47TP4DECN4R/events.json","paper":"https://pith.science/paper/5S2T7NLD"},"agent_actions":{"view_html":"https://pith.science/pith/5S2T7NLDWNND7OO47TP4DECN4R","download_json":"https://pith.science/pith/5S2T7NLDWNND7OO47TP4DECN4R.json","view_paper":"https://pith.science/paper/5S2T7NLD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2007.13828&json=true","fetch_graph":"https://pith.science/api/pith-number/5S2T7NLDWNND7OO47TP4DECN4R/graph.json","fetch_events":"https://pith.science/api/pith-number/5S2T7NLDWNND7OO47TP4DECN4R/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5S2T7NLDWNND7OO47TP4DECN4R/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5S2T7NLDWNND7OO47TP4DECN4R/action/storage_attestation","attest_author":"https://pith.science/pith/5S2T7NLDWNND7OO47TP4DECN4R/action/author_attestation","sign_citation":"https://pith.science/pith/5S2T7NLDWNND7OO47TP4DECN4R/action/citation_signature","submit_replication":"https://pith.science/pith/5S2T7NLDWNND7OO47TP4DECN4R/action/replication_record"}},"created_at":"2026-07-05T01:23:26.110161+00:00","updated_at":"2026-07-05T01:23:26.110161+00:00"}