{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:22GR526EC75QM7QTL4JJRG4B7Z","short_pith_number":"pith:22GR526E","canonical_record":{"source":{"id":"1804.10223","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NE","submitted_at":"2018-04-26T18:18:57Z","cross_cats_sorted":["cs.DC","cs.LG"],"title_canon_sha256":"48ff814c42567cf9e682917ec0e65690c9b6eff5624406de8dc7383bc8b2da7e","abstract_canon_sha256":"becccc62ff311111b4d2493055ae8e7b6170a1641433ee93ec332a152873f09e"},"schema_version":"1.0"},"canonical_sha256":"d68d1eebc417fb067e135f12989b81fe480728f704ed4ee8af9f18383639261d","source":{"kind":"arxiv","id":"1804.10223","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1804.10223","created_at":"2026-05-18T00:17:21Z"},{"alias_kind":"arxiv_version","alias_value":"1804.10223v1","created_at":"2026-05-18T00:17:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1804.10223","created_at":"2026-05-18T00:17:21Z"},{"alias_kind":"pith_short_12","alias_value":"22GR526EC75Q","created_at":"2026-05-18T12:31:59Z"},{"alias_kind":"pith_short_16","alias_value":"22GR526EC75QM7QT","created_at":"2026-05-18T12:31:59Z"},{"alias_kind":"pith_short_8","alias_value":"22GR526E","created_at":"2026-05-18T12:31:59Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:22GR526EC75QM7QTL4JJRG4B7Z","target":"record","payload":{"canonical_record":{"source":{"id":"1804.10223","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NE","submitted_at":"2018-04-26T18:18:57Z","cross_cats_sorted":["cs.DC","cs.LG"],"title_canon_sha256":"48ff814c42567cf9e682917ec0e65690c9b6eff5624406de8dc7383bc8b2da7e","abstract_canon_sha256":"becccc62ff311111b4d2493055ae8e7b6170a1641433ee93ec332a152873f09e"},"schema_version":"1.0"},"canonical_sha256":"d68d1eebc417fb067e135f12989b81fe480728f704ed4ee8af9f18383639261d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:17:21.416642Z","signature_b64":"MBPzeZA7xIi+1KXxfjNjxN3IKYmDLbJVT8NJmkPhf/g/lsI2irpVtQUjYEMeKB2/tOri+q91wpkdmUkGLEBoBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d68d1eebc417fb067e135f12989b81fe480728f704ed4ee8af9f18383639261d","last_reissued_at":"2026-05-18T00:17:21.415905Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:17:21.415905Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1804.10223","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:17:21Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TzHXfiWugThYtSuKLwvwMCNdQ760/7XBDLSkZhlLtwxGEsTo5VDzpwn9nu/IX37/mA7BjxwCR05OjZGaAadnDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-30T06:14:09.621943Z"},"content_sha256":"f929cf92812ba53c4b880ea6358f7dcdec755203c59ab1c0904779853b6fbeaf","schema_version":"1.0","event_id":"sha256:f929cf92812ba53c4b880ea6358f7dcdec755203c59ab1c0904779853b6fbeaf"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:22GR526EC75QM7QTL4JJRG4B7Z","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Sparse Persistent RNNs: Squeezing Large Recurrent Networks On-Chip","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC","cs.LG"],"primary_cat":"cs.NE","authors_text":"Feiwen Zhu, Fung Xie, Jeff Pool, Jeremy Appleyard, Michael Andersch","submitted_at":"2018-04-26T18:18:57Z","abstract_excerpt":"Recurrent Neural Networks (RNNs) are powerful tools for solving sequence-based problems, but their efficacy and execution time are dependent on the size of the network. Following recent work in simplifying these networks with model pruning and a novel mapping of work onto GPUs, we design an efficient implementation for sparse RNNs. We investigate several optimizations and tradeoffs: Lamport timestamps, wide memory loads, and a bank-aware weight layout. With these optimizations, we achieve speedups of over 6x over the next best algorithm for a hidden layer of size 2304, batch size of 4, and a d"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1804.10223","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":""},"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:17:21Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"B65N7vmaHyjYghyLq3qTmxO0tnZ9nWjuCTQ5OG9zOEEIbe237s2HdYh7jy4lt60tbtVmWMgNIssest7D6pgjAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-30T06:14:09.622359Z"},"content_sha256":"a9f98c631944e23d0286f73a9b52470da8ecaa850f66995009bf8894c41c18ee","schema_version":"1.0","event_id":"sha256:a9f98c631944e23d0286f73a9b52470da8ecaa850f66995009bf8894c41c18ee"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/22GR526EC75QM7QTL4JJRG4B7Z/bundle.json","state_url":"https://pith.science/pith/22GR526EC75QM7QTL4JJRG4B7Z/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/22GR526EC75QM7QTL4JJRG4B7Z/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-05-30T06:14:09Z","links":{"resolver":"https://pith.science/pith/22GR526EC75QM7QTL4JJRG4B7Z","bundle":"https://pith.science/pith/22GR526EC75QM7QTL4JJRG4B7Z/bundle.json","state":"https://pith.science/pith/22GR526EC75QM7QTL4JJRG4B7Z/state.json","well_known_bundle":"https://pith.science/.well-known/pith/22GR526EC75QM7QTL4JJRG4B7Z/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:22GR526EC75QM7QTL4JJRG4B7Z","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"becccc62ff311111b4d2493055ae8e7b6170a1641433ee93ec332a152873f09e","cross_cats_sorted":["cs.DC","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NE","submitted_at":"2018-04-26T18:18:57Z","title_canon_sha256":"48ff814c42567cf9e682917ec0e65690c9b6eff5624406de8dc7383bc8b2da7e"},"schema_version":"1.0","source":{"id":"1804.10223","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1804.10223","created_at":"2026-05-18T00:17:21Z"},{"alias_kind":"arxiv_version","alias_value":"1804.10223v1","created_at":"2026-05-18T00:17:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1804.10223","created_at":"2026-05-18T00:17:21Z"},{"alias_kind":"pith_short_12","alias_value":"22GR526EC75Q","created_at":"2026-05-18T12:31:59Z"},{"alias_kind":"pith_short_16","alias_value":"22GR526EC75QM7QT","created_at":"2026-05-18T12:31:59Z"},{"alias_kind":"pith_short_8","alias_value":"22GR526E","created_at":"2026-05-18T12:31:59Z"}],"graph_snapshots":[{"event_id":"sha256:a9f98c631944e23d0286f73a9b52470da8ecaa850f66995009bf8894c41c18ee","target":"graph","created_at":"2026-05-18T00:17:21Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"Recurrent Neural Networks (RNNs) are powerful tools for solving sequence-based problems, but their efficacy and execution time are dependent on the size of the network. Following recent work in simplifying these networks with model pruning and a novel mapping of work onto GPUs, we design an efficient implementation for sparse RNNs. We investigate several optimizations and tradeoffs: Lamport timestamps, wide memory loads, and a bank-aware weight layout. With these optimizations, we achieve speedups of over 6x over the next best algorithm for a hidden layer of size 2304, batch size of 4, and a d","authors_text":"Feiwen Zhu, Fung Xie, Jeff Pool, Jeremy Appleyard, Michael Andersch","cross_cats":["cs.DC","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NE","submitted_at":"2018-04-26T18:18:57Z","title":"Sparse Persistent RNNs: Squeezing Large Recurrent Networks On-Chip"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1804.10223","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:f929cf92812ba53c4b880ea6358f7dcdec755203c59ab1c0904779853b6fbeaf","target":"record","created_at":"2026-05-18T00:17:21Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"becccc62ff311111b4d2493055ae8e7b6170a1641433ee93ec332a152873f09e","cross_cats_sorted":["cs.DC","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NE","submitted_at":"2018-04-26T18:18:57Z","title_canon_sha256":"48ff814c42567cf9e682917ec0e65690c9b6eff5624406de8dc7383bc8b2da7e"},"schema_version":"1.0","source":{"id":"1804.10223","kind":"arxiv","version":1}},"canonical_sha256":"d68d1eebc417fb067e135f12989b81fe480728f704ed4ee8af9f18383639261d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d68d1eebc417fb067e135f12989b81fe480728f704ed4ee8af9f18383639261d","first_computed_at":"2026-05-18T00:17:21.415905Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:17:21.415905Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"MBPzeZA7xIi+1KXxfjNjxN3IKYmDLbJVT8NJmkPhf/g/lsI2irpVtQUjYEMeKB2/tOri+q91wpkdmUkGLEBoBQ==","signature_status":"signed_v1","signed_at":"2026-05-18T00:17:21.416642Z","signed_message":"canonical_sha256_bytes"},"source_id":"1804.10223","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f929cf92812ba53c4b880ea6358f7dcdec755203c59ab1c0904779853b6fbeaf","sha256:a9f98c631944e23d0286f73a9b52470da8ecaa850f66995009bf8894c41c18ee"],"state_sha256":"6f539684d0bd1b0cb63214c0c7983bb5b140aea46e9da93237518e580cf7f340"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ZPW9miE159Czfmqf3ofnViXdlCqiZDJP70Yy1GGbPRC+OjcXOzK8BRmtxnmtjnCel/e1D2vJW9CnXXOUjAjGDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-30T06:14:09.624342Z","bundle_sha256":"0d04f987306f34bbabd7ec38523df2f2009df6062aa37598a3e78fdae24ecb77"}}