{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:3NJKI7MESPM4HJ2DIMUW4MDN6D","short_pith_number":"pith:3NJKI7ME","schema_version":"1.0","canonical_sha256":"db52a47d8493d9c3a74343296e306df0e49a7ebcb656b7d5906136a8f32b72bc","source":{"kind":"arxiv","id":"2312.06532","version":3},"attestation_state":"computed","paper":{"title":"RACE-IT: A Reconfigurable Analog Computing Engine for In-Memory Transformer Acceleration","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.ET","cs.LG"],"primary_cat":"cs.AR","authors_text":"Aishwarya Natarajan, Archit Gajjar, Giacomo Pedretti, Jim Ignowski, John Moon, Lei Zhao, Luca Buonanno, Ron M. Roth, Sergey Serebryakov","submitted_at":"2023-11-29T22:45:39Z","abstract_excerpt":"Transformer models represent the cutting edge of Deep Neural Networks (DNNs) and excel in a wide range of machine learning tasks. However, processing these models demands significant computational resources and results in a substantial memory footprint. While In-memory Computing (IMC)offers promise for accelerating Vector-Matrix Multiplications(VMMs) with high computational parallelism and minimal data movement, employing it for other crucial DNN operators remains a formidable task. This challenge is exacerbated by the extensive use of complex activation functions, Softmax, and data-dependent "},"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":"2312.06532","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.AR","submitted_at":"2023-11-29T22:45:39Z","cross_cats_sorted":["cs.ET","cs.LG"],"title_canon_sha256":"8ee47ec147a406f198490cb408b0aafb7f129a81a6085cc0c3032e5fd036812e","abstract_canon_sha256":"1cdd4a8e6aec6d289626ee9defefd9b6962e46b3af364275d9b5df6d812a6d93"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:49:44.883242Z","signature_b64":"Mo1wrJjIhgVpQPFVHLLBakPVkg+qf8wo7GsXS3w8wtF0AaUHmhmgr0BC1EEEktaIGlhyt898cGBBqJ5/PaJYDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"db52a47d8493d9c3a74343296e306df0e49a7ebcb656b7d5906136a8f32b72bc","last_reissued_at":"2026-07-05T11:49:44.882787Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:49:44.882787Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RACE-IT: A Reconfigurable Analog Computing Engine for In-Memory Transformer Acceleration","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.ET","cs.LG"],"primary_cat":"cs.AR","authors_text":"Aishwarya Natarajan, Archit Gajjar, Giacomo Pedretti, Jim Ignowski, John Moon, Lei Zhao, Luca Buonanno, Ron M. Roth, Sergey Serebryakov","submitted_at":"2023-11-29T22:45:39Z","abstract_excerpt":"Transformer models represent the cutting edge of Deep Neural Networks (DNNs) and excel in a wide range of machine learning tasks. However, processing these models demands significant computational resources and results in a substantial memory footprint. While In-memory Computing (IMC)offers promise for accelerating Vector-Matrix Multiplications(VMMs) with high computational parallelism and minimal data movement, employing it for other crucial DNN operators remains a formidable task. This challenge is exacerbated by the extensive use of complex activation functions, Softmax, and data-dependent "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.06532","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/2312.06532/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":"2312.06532","created_at":"2026-07-05T11:49:44.882840+00:00"},{"alias_kind":"arxiv_version","alias_value":"2312.06532v3","created_at":"2026-07-05T11:49:44.882840+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.06532","created_at":"2026-07-05T11:49:44.882840+00:00"},{"alias_kind":"pith_short_12","alias_value":"3NJKI7MESPM4","created_at":"2026-07-05T11:49:44.882840+00:00"},{"alias_kind":"pith_short_16","alias_value":"3NJKI7MESPM4HJ2D","created_at":"2026-07-05T11:49:44.882840+00:00"},{"alias_kind":"pith_short_8","alias_value":"3NJKI7ME","created_at":"2026-07-05T11:49:44.882840+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.11847","citing_title":"A Fast and Energy-Efficient Latch-Based Memristive Analog Content-Addressable Memory","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3NJKI7MESPM4HJ2DIMUW4MDN6D","json":"https://pith.science/pith/3NJKI7MESPM4HJ2DIMUW4MDN6D.json","graph_json":"https://pith.science/api/pith-number/3NJKI7MESPM4HJ2DIMUW4MDN6D/graph.json","events_json":"https://pith.science/api/pith-number/3NJKI7MESPM4HJ2DIMUW4MDN6D/events.json","paper":"https://pith.science/paper/3NJKI7ME"},"agent_actions":{"view_html":"https://pith.science/pith/3NJKI7MESPM4HJ2DIMUW4MDN6D","download_json":"https://pith.science/pith/3NJKI7MESPM4HJ2DIMUW4MDN6D.json","view_paper":"https://pith.science/paper/3NJKI7ME","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2312.06532&json=true","fetch_graph":"https://pith.science/api/pith-number/3NJKI7MESPM4HJ2DIMUW4MDN6D/graph.json","fetch_events":"https://pith.science/api/pith-number/3NJKI7MESPM4HJ2DIMUW4MDN6D/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3NJKI7MESPM4HJ2DIMUW4MDN6D/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3NJKI7MESPM4HJ2DIMUW4MDN6D/action/storage_attestation","attest_author":"https://pith.science/pith/3NJKI7MESPM4HJ2DIMUW4MDN6D/action/author_attestation","sign_citation":"https://pith.science/pith/3NJKI7MESPM4HJ2DIMUW4MDN6D/action/citation_signature","submit_replication":"https://pith.science/pith/3NJKI7MESPM4HJ2DIMUW4MDN6D/action/replication_record"}},"created_at":"2026-07-05T11:49:44.882840+00:00","updated_at":"2026-07-05T11:49:44.882840+00:00"}