{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:EVM56RF6OSQHKJPCDKNTCTDXQD","short_pith_number":"pith:EVM56RF6","schema_version":"1.0","canonical_sha256":"2559df44be74a07525e21a9b314c7780c44e771b877d7ab73d8ccd0413fcd5af","source":{"kind":"arxiv","id":"1908.08976","version":1},"attestation_state":"computed","paper":{"title":"MASR: A Modular Accelerator for Sparse RNNs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"eess.SP","authors_text":"Alexander M. Rush, Brandon Reagen, David Brooks, Gu-Yeon Wei, Lillian Pentecost, Marco Donato, Thierry Tambe, Udit Gupta","submitted_at":"2019-08-23T18:58:07Z","abstract_excerpt":"Recurrent neural networks (RNNs) are becoming the de facto solution for speech recognition. RNNs exploit long-term temporal relationships in data by applying repeated, learned transformations. Unlike fully-connected (FC) layers with single vector matrix operations, RNN layers consist of hundreds of such operations chained over time. This poses challenges unique to RNNs that are not found in convolutional neural networks (CNNs) or FC models, namely large dynamic activation. In this paper we present MASR, a principled and modular architecture that accelerates bidirectional RNNs for on-chip ASR. "},"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":"1908.08976","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.SP","submitted_at":"2019-08-23T18:58:07Z","cross_cats_sorted":[],"title_canon_sha256":"60fd1ac93684be9924e434ab461c2f3f83940fa00eab24b21811cbf4de031b99","abstract_canon_sha256":"01130b12afcb6b4b24ba014c58a841feee4ddd0d225625e1db8a5563a2198880"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-04T23:59:35.378438Z","signature_b64":"3cJMCjvnn/Wa6y++nk6Rv7I4YqgZFeQvxLwRo4jecp22G7W0MjYNjattEgpC/VOS11BsQ6iUj/6LPZIZ+CMTBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2559df44be74a07525e21a9b314c7780c44e771b877d7ab73d8ccd0413fcd5af","last_reissued_at":"2026-07-04T23:59:35.377948Z","signature_status":"signed_v1","first_computed_at":"2026-07-04T23:59:35.377948Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MASR: A Modular Accelerator for Sparse RNNs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"eess.SP","authors_text":"Alexander M. Rush, Brandon Reagen, David Brooks, Gu-Yeon Wei, Lillian Pentecost, Marco Donato, Thierry Tambe, Udit Gupta","submitted_at":"2019-08-23T18:58:07Z","abstract_excerpt":"Recurrent neural networks (RNNs) are becoming the de facto solution for speech recognition. RNNs exploit long-term temporal relationships in data by applying repeated, learned transformations. Unlike fully-connected (FC) layers with single vector matrix operations, RNN layers consist of hundreds of such operations chained over time. This poses challenges unique to RNNs that are not found in convolutional neural networks (CNNs) or FC models, namely large dynamic activation. In this paper we present MASR, a principled and modular architecture that accelerates bidirectional RNNs for on-chip ASR. "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1908.08976","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/1908.08976/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":"1908.08976","created_at":"2026-07-04T23:59:35.378009+00:00"},{"alias_kind":"arxiv_version","alias_value":"1908.08976v1","created_at":"2026-07-04T23:59:35.378009+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1908.08976","created_at":"2026-07-04T23:59:35.378009+00:00"},{"alias_kind":"pith_short_12","alias_value":"EVM56RF6OSQH","created_at":"2026-07-04T23:59:35.378009+00:00"},{"alias_kind":"pith_short_16","alias_value":"EVM56RF6OSQHKJPC","created_at":"2026-07-04T23:59:35.378009+00:00"},{"alias_kind":"pith_short_8","alias_value":"EVM56RF6","created_at":"2026-07-04T23:59:35.378009+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/EVM56RF6OSQHKJPCDKNTCTDXQD","json":"https://pith.science/pith/EVM56RF6OSQHKJPCDKNTCTDXQD.json","graph_json":"https://pith.science/api/pith-number/EVM56RF6OSQHKJPCDKNTCTDXQD/graph.json","events_json":"https://pith.science/api/pith-number/EVM56RF6OSQHKJPCDKNTCTDXQD/events.json","paper":"https://pith.science/paper/EVM56RF6"},"agent_actions":{"view_html":"https://pith.science/pith/EVM56RF6OSQHKJPCDKNTCTDXQD","download_json":"https://pith.science/pith/EVM56RF6OSQHKJPCDKNTCTDXQD.json","view_paper":"https://pith.science/paper/EVM56RF6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1908.08976&json=true","fetch_graph":"https://pith.science/api/pith-number/EVM56RF6OSQHKJPCDKNTCTDXQD/graph.json","fetch_events":"https://pith.science/api/pith-number/EVM56RF6OSQHKJPCDKNTCTDXQD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EVM56RF6OSQHKJPCDKNTCTDXQD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EVM56RF6OSQHKJPCDKNTCTDXQD/action/storage_attestation","attest_author":"https://pith.science/pith/EVM56RF6OSQHKJPCDKNTCTDXQD/action/author_attestation","sign_citation":"https://pith.science/pith/EVM56RF6OSQHKJPCDKNTCTDXQD/action/citation_signature","submit_replication":"https://pith.science/pith/EVM56RF6OSQHKJPCDKNTCTDXQD/action/replication_record"}},"created_at":"2026-07-04T23:59:35.378009+00:00","updated_at":"2026-07-04T23:59:35.378009+00:00"}