{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2016:OSIJ76O2D4AFZCONK4OW3E2LUP","short_pith_number":"pith:OSIJ76O2","canonical_record":{"source":{"id":"1604.01946","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-04-07T10:31:01Z","cross_cats_sorted":["cs.NE"],"title_canon_sha256":"7e5236f19e1c3c4e43f43bd50b374712c41aac10df20c127ab8f9ff9cbea1266","abstract_canon_sha256":"7c6115ac3a6d6e2e25c7c7939406bcafe6ffb42d097813a083397529b5f55ff7"},"schema_version":"1.0"},"canonical_sha256":"74909ff9da1f005c89cd571d6d934ba3d025020f7e9c088f8a48665c01c84a41","source":{"kind":"arxiv","id":"1604.01946","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1604.01946","created_at":"2026-05-18T01:17:31Z"},{"alias_kind":"arxiv_version","alias_value":"1604.01946v1","created_at":"2026-05-18T01:17:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1604.01946","created_at":"2026-05-18T01:17:31Z"},{"alias_kind":"pith_short_12","alias_value":"OSIJ76O2D4AF","created_at":"2026-05-18T12:30:36Z"},{"alias_kind":"pith_short_16","alias_value":"OSIJ76O2D4AFZCON","created_at":"2026-05-18T12:30:36Z"},{"alias_kind":"pith_short_8","alias_value":"OSIJ76O2","created_at":"2026-05-18T12:30:36Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2016:OSIJ76O2D4AFZCONK4OW3E2LUP","target":"record","payload":{"canonical_record":{"source":{"id":"1604.01946","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-04-07T10:31:01Z","cross_cats_sorted":["cs.NE"],"title_canon_sha256":"7e5236f19e1c3c4e43f43bd50b374712c41aac10df20c127ab8f9ff9cbea1266","abstract_canon_sha256":"7c6115ac3a6d6e2e25c7c7939406bcafe6ffb42d097813a083397529b5f55ff7"},"schema_version":"1.0"},"canonical_sha256":"74909ff9da1f005c89cd571d6d934ba3d025020f7e9c088f8a48665c01c84a41","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T01:17:31.630124Z","signature_b64":"LJ8OfnZ2Hku7YOUdKjoOGhmtGGJJVjxa3bUqn91l4aocGaXTHvGr7ljWWVVhMQGXNKAQZQq9oRhUSC8Q2S2qBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"74909ff9da1f005c89cd571d6d934ba3d025020f7e9c088f8a48665c01c84a41","last_reissued_at":"2026-05-18T01:17:31.629367Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T01:17:31.629367Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1604.01946","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":"2026-05-18T01:17:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6J6+FaqD7OXITvXHuze5tw81CAv1dDPatCIV6RYq76EWRMGlmCRXs704m5IEHLfLjJscmYvkB0iOkjqDHREtBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-04T17:38:32.888434Z"},"content_sha256":"60d5140b3249665daebb205bfb6fdf35b5c02d74ade1f1327c2fbaaaf3389723","schema_version":"1.0","event_id":"sha256:60d5140b3249665daebb205bfb6fdf35b5c02d74ade1f1327c2fbaaaf3389723"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2016:OSIJ76O2D4AFZCONK4OW3E2LUP","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Optimizing Performance of Recurrent Neural Networks on GPUs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.NE"],"primary_cat":"cs.LG","authors_text":"Jeremy Appleyard, Phil Blunsom, Tomas Kocisky","submitted_at":"2016-04-07T10:31:01Z","abstract_excerpt":"As recurrent neural networks become larger and deeper, training times for single networks are rising into weeks or even months. As such there is a significant incentive to improve the performance and scalability of these networks. While GPUs have become the hardware of choice for training and deploying recurrent models, the implementations employed often make use of only basic optimizations for these architectures. In this article we demonstrate that by exposing parallelism between operations within the network, an order of magnitude speedup across a range of network sizes can be achieved over"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1604.01946","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":"2026-05-18T01:17:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"cXXM5CWoVQja2iPBmnAtLkWAjP2rtt/tppoBDs6L3ONH0i58e3laCMSzgxkefInYCc33XUWcr0Eb1X2yXwzaCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-04T17:38:32.889039Z"},"content_sha256":"901c643964de99fd1c85633ec7d01c7ea7fa9b7306d14a94395742570e152b08","schema_version":"1.0","event_id":"sha256:901c643964de99fd1c85633ec7d01c7ea7fa9b7306d14a94395742570e152b08"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/OSIJ76O2D4AFZCONK4OW3E2LUP/bundle.json","state_url":"https://pith.science/pith/OSIJ76O2D4AFZCONK4OW3E2LUP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/OSIJ76O2D4AFZCONK4OW3E2LUP/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":"2026-06-04T17:38:32Z","links":{"resolver":"https://pith.science/pith/OSIJ76O2D4AFZCONK4OW3E2LUP","bundle":"https://pith.science/pith/OSIJ76O2D4AFZCONK4OW3E2LUP/bundle.json","state":"https://pith.science/pith/OSIJ76O2D4AFZCONK4OW3E2LUP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/OSIJ76O2D4AFZCONK4OW3E2LUP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2016:OSIJ76O2D4AFZCONK4OW3E2LUP","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":"7c6115ac3a6d6e2e25c7c7939406bcafe6ffb42d097813a083397529b5f55ff7","cross_cats_sorted":["cs.NE"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-04-07T10:31:01Z","title_canon_sha256":"7e5236f19e1c3c4e43f43bd50b374712c41aac10df20c127ab8f9ff9cbea1266"},"schema_version":"1.0","source":{"id":"1604.01946","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1604.01946","created_at":"2026-05-18T01:17:31Z"},{"alias_kind":"arxiv_version","alias_value":"1604.01946v1","created_at":"2026-05-18T01:17:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1604.01946","created_at":"2026-05-18T01:17:31Z"},{"alias_kind":"pith_short_12","alias_value":"OSIJ76O2D4AF","created_at":"2026-05-18T12:30:36Z"},{"alias_kind":"pith_short_16","alias_value":"OSIJ76O2D4AFZCON","created_at":"2026-05-18T12:30:36Z"},{"alias_kind":"pith_short_8","alias_value":"OSIJ76O2","created_at":"2026-05-18T12:30:36Z"}],"graph_snapshots":[{"event_id":"sha256:901c643964de99fd1c85633ec7d01c7ea7fa9b7306d14a94395742570e152b08","target":"graph","created_at":"2026-05-18T01:17:31Z","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":"As recurrent neural networks become larger and deeper, training times for single networks are rising into weeks or even months. As such there is a significant incentive to improve the performance and scalability of these networks. While GPUs have become the hardware of choice for training and deploying recurrent models, the implementations employed often make use of only basic optimizations for these architectures. In this article we demonstrate that by exposing parallelism between operations within the network, an order of magnitude speedup across a range of network sizes can be achieved over","authors_text":"Jeremy Appleyard, Phil Blunsom, Tomas Kocisky","cross_cats":["cs.NE"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-04-07T10:31:01Z","title":"Optimizing Performance of Recurrent Neural Networks on GPUs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1604.01946","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":"sha256:60d5140b3249665daebb205bfb6fdf35b5c02d74ade1f1327c2fbaaaf3389723","target":"record","created_at":"2026-05-18T01:17:31Z","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":"7c6115ac3a6d6e2e25c7c7939406bcafe6ffb42d097813a083397529b5f55ff7","cross_cats_sorted":["cs.NE"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-04-07T10:31:01Z","title_canon_sha256":"7e5236f19e1c3c4e43f43bd50b374712c41aac10df20c127ab8f9ff9cbea1266"},"schema_version":"1.0","source":{"id":"1604.01946","kind":"arxiv","version":1}},"canonical_sha256":"74909ff9da1f005c89cd571d6d934ba3d025020f7e9c088f8a48665c01c84a41","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"74909ff9da1f005c89cd571d6d934ba3d025020f7e9c088f8a48665c01c84a41","first_computed_at":"2026-05-18T01:17:31.629367Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T01:17:31.629367Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"LJ8OfnZ2Hku7YOUdKjoOGhmtGGJJVjxa3bUqn91l4aocGaXTHvGr7ljWWVVhMQGXNKAQZQq9oRhUSC8Q2S2qBg==","signature_status":"signed_v1","signed_at":"2026-05-18T01:17:31.630124Z","signed_message":"canonical_sha256_bytes"},"source_id":"1604.01946","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:60d5140b3249665daebb205bfb6fdf35b5c02d74ade1f1327c2fbaaaf3389723","sha256:901c643964de99fd1c85633ec7d01c7ea7fa9b7306d14a94395742570e152b08"],"state_sha256":"ed1580c418fadc9c43a1e4b197c3e57a2e82a84307a68898b955e36b99211245"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"IQRhd9nNHyFUSK7BG1YZ+zLQf04nY1KkvgFEuNF8AI66oI+cNtxyDs/aUNLgfV6pUi6e8dHMte2WUYuR6a4GDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-04T17:38:32.891871Z","bundle_sha256":"019456a4286533627ed68a3340588bfaf39ac4fedb8df8639a577292db615852"}}