{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:MMMFAOU554YXK5VUR4VS4RCSS5","short_pith_number":"pith:MMMFAOU5","schema_version":"1.0","canonical_sha256":"6318503a9def317576b48f2b2e445297701d601c758619937820e16c24b37983","source":{"kind":"arxiv","id":"2108.07879","version":1},"attestation_state":"computed","paper":{"title":"Edge AI without Compromise: Efficient, Versatile and Accurate Neurocomputing in Resistive Random-Access Memory","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.ET","cs.LG"],"primary_cat":"cs.AR","authors_text":"2), (2) University of California San Diego, (3) Tsinghua University, (4) University of Notre Dame, 5), (5) University of Pittsburgh), Bin Gao (3), Clemens Schaefer (4), Dabin Wu (3), Gert Cauwenberghs (2) ((1) Stanford University, He Qian (3), H.-S. Philip Wong (1), Huaqiang Wu (3), Priyanka Raina (1), Rajkumar Kubendran (2, S. Burc Eryilmaz (1), Siddharth Joshi (4, Stephen Deiss (2), Weier Wan (1), Wenqiang Zhang (3)","submitted_at":"2021-08-17T21:08:51Z","abstract_excerpt":"Realizing today's cloud-level artificial intelligence functionalities directly on devices distributed at the edge of the internet calls for edge hardware capable of processing multiple modalities of sensory data (e.g. video, audio) at unprecedented energy-efficiency. AI hardware architectures today cannot meet the demand due to a fundamental \"memory wall\": data movement between separate compute and memory units consumes large energy and incurs long latency. Resistive random-access memory (RRAM) based compute-in-memory (CIM) architectures promise to bring orders of magnitude energy-efficiency i"},"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":"2108.07879","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AR","submitted_at":"2021-08-17T21:08:51Z","cross_cats_sorted":["cs.AI","cs.ET","cs.LG"],"title_canon_sha256":"84b88f92d980fb46467c3f009eb9302ca66451e19480016c73806b6f87c36815","abstract_canon_sha256":"71973a08c7914106f58de8a993795e74d2dd24e0483e44050a6f733da22ad8dc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:06:49.071077Z","signature_b64":"NgTLsjwyTmUxFgcAFGgjhBOu8mfTgoFpLH5hrdFDoRuRuk4dRqjjWAyZz8FSvXxuEgXP+WNeV4ZiCSAyQ5wbAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6318503a9def317576b48f2b2e445297701d601c758619937820e16c24b37983","last_reissued_at":"2026-07-05T03:06:49.070679Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:06:49.070679Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Edge AI without Compromise: Efficient, Versatile and Accurate Neurocomputing in Resistive Random-Access Memory","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.ET","cs.LG"],"primary_cat":"cs.AR","authors_text":"2), (2) University of California San Diego, (3) Tsinghua University, (4) University of Notre Dame, 5), (5) University of Pittsburgh), Bin Gao (3), Clemens Schaefer (4), Dabin Wu (3), Gert Cauwenberghs (2) ((1) Stanford University, He Qian (3), H.-S. Philip Wong (1), Huaqiang Wu (3), Priyanka Raina (1), Rajkumar Kubendran (2, S. Burc Eryilmaz (1), Siddharth Joshi (4, Stephen Deiss (2), Weier Wan (1), Wenqiang Zhang (3)","submitted_at":"2021-08-17T21:08:51Z","abstract_excerpt":"Realizing today's cloud-level artificial intelligence functionalities directly on devices distributed at the edge of the internet calls for edge hardware capable of processing multiple modalities of sensory data (e.g. video, audio) at unprecedented energy-efficiency. AI hardware architectures today cannot meet the demand due to a fundamental \"memory wall\": data movement between separate compute and memory units consumes large energy and incurs long latency. Resistive random-access memory (RRAM) based compute-in-memory (CIM) architectures promise to bring orders of magnitude energy-efficiency i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2108.07879","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/2108.07879/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":"2108.07879","created_at":"2026-07-05T03:06:49.070738+00:00"},{"alias_kind":"arxiv_version","alias_value":"2108.07879v1","created_at":"2026-07-05T03:06:49.070738+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2108.07879","created_at":"2026-07-05T03:06:49.070738+00:00"},{"alias_kind":"pith_short_12","alias_value":"MMMFAOU554YX","created_at":"2026-07-05T03:06:49.070738+00:00"},{"alias_kind":"pith_short_16","alias_value":"MMMFAOU554YXK5VU","created_at":"2026-07-05T03:06:49.070738+00:00"},{"alias_kind":"pith_short_8","alias_value":"MMMFAOU5","created_at":"2026-07-05T03:06:49.070738+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/MMMFAOU554YXK5VUR4VS4RCSS5","json":"https://pith.science/pith/MMMFAOU554YXK5VUR4VS4RCSS5.json","graph_json":"https://pith.science/api/pith-number/MMMFAOU554YXK5VUR4VS4RCSS5/graph.json","events_json":"https://pith.science/api/pith-number/MMMFAOU554YXK5VUR4VS4RCSS5/events.json","paper":"https://pith.science/paper/MMMFAOU5"},"agent_actions":{"view_html":"https://pith.science/pith/MMMFAOU554YXK5VUR4VS4RCSS5","download_json":"https://pith.science/pith/MMMFAOU554YXK5VUR4VS4RCSS5.json","view_paper":"https://pith.science/paper/MMMFAOU5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2108.07879&json=true","fetch_graph":"https://pith.science/api/pith-number/MMMFAOU554YXK5VUR4VS4RCSS5/graph.json","fetch_events":"https://pith.science/api/pith-number/MMMFAOU554YXK5VUR4VS4RCSS5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MMMFAOU554YXK5VUR4VS4RCSS5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MMMFAOU554YXK5VUR4VS4RCSS5/action/storage_attestation","attest_author":"https://pith.science/pith/MMMFAOU554YXK5VUR4VS4RCSS5/action/author_attestation","sign_citation":"https://pith.science/pith/MMMFAOU554YXK5VUR4VS4RCSS5/action/citation_signature","submit_replication":"https://pith.science/pith/MMMFAOU554YXK5VUR4VS4RCSS5/action/replication_record"}},"created_at":"2026-07-05T03:06:49.070738+00:00","updated_at":"2026-07-05T03:06:49.070738+00:00"}