{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2015:75G6SZWEUWEIL74MBM376JGOXI","short_pith_number":"pith:75G6SZWE","canonical_record":{"source":{"id":"1503.02852","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NE","submitted_at":"2015-03-10T10:27:55Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"10df6b001e9050eadd85820c536a12615d03547287ef9289e02d76c5c427e865","abstract_canon_sha256":"a6fe71ebcf47c43c74ce381d914b11b2462b6ae5727d613e76a3c80dbffd1fbf"},"schema_version":"1.0"},"canonical_sha256":"ff4de966c4a58885ff8c0b37ff24ceba27e0d0cb5bce65bf95a069c8b887fc1d","source":{"kind":"arxiv","id":"1503.02852","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1503.02852","created_at":"2026-05-18T01:26:10Z"},{"alias_kind":"arxiv_version","alias_value":"1503.02852v1","created_at":"2026-05-18T01:26:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1503.02852","created_at":"2026-05-18T01:26:10Z"},{"alias_kind":"pith_short_12","alias_value":"75G6SZWEUWEI","created_at":"2026-05-18T12:29:07Z"},{"alias_kind":"pith_short_16","alias_value":"75G6SZWEUWEIL74M","created_at":"2026-05-18T12:29:07Z"},{"alias_kind":"pith_short_8","alias_value":"75G6SZWE","created_at":"2026-05-18T12:29:07Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2015:75G6SZWEUWEIL74MBM376JGOXI","target":"record","payload":{"canonical_record":{"source":{"id":"1503.02852","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NE","submitted_at":"2015-03-10T10:27:55Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"10df6b001e9050eadd85820c536a12615d03547287ef9289e02d76c5c427e865","abstract_canon_sha256":"a6fe71ebcf47c43c74ce381d914b11b2462b6ae5727d613e76a3c80dbffd1fbf"},"schema_version":"1.0"},"canonical_sha256":"ff4de966c4a58885ff8c0b37ff24ceba27e0d0cb5bce65bf95a069c8b887fc1d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T01:26:10.544354Z","signature_b64":"TsJciua/VBHJRSsvxtMhKVisAlXgCIPdiAatw8htkDxQjgCV64MVTwwpBrJ0T49aFhuOhh+GVS73+2v1ImZeCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ff4de966c4a58885ff8c0b37ff24ceba27e0d0cb5bce65bf95a069c8b887fc1d","last_reissued_at":"2026-05-18T01:26:10.543752Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T01:26:10.543752Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1503.02852","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":"2026-05-18T01:26:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"j/3LcsXD6hGRFgsPnzZZ61JNhlkwOLrwejvyAEvjgXMPAib88sgj25EGS+gQG5uH44YP+Ar3++S4nxurFGynBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-11T22:47:56.518438Z"},"content_sha256":"9f9d58ee8b60a2ba8f6fb69fd490758ad7c5e8aa80fe188646ffcb56bf48e6c2","schema_version":"1.0","event_id":"sha256:9f9d58ee8b60a2ba8f6fb69fd490758ad7c5e8aa80fe188646ffcb56bf48e6c2"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2015:75G6SZWEUWEIL74MBM376JGOXI","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Single stream parallelization of generalized LSTM-like RNNs on a GPU","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.NE","authors_text":"Kyuyeon Hwang, Wonyong Sung","submitted_at":"2015-03-10T10:27:55Z","abstract_excerpt":"Recurrent neural networks (RNNs) have shown outstanding performance on processing sequence data. However, they suffer from long training time, which demands parallel implementations of the training procedure. Parallelization of the training algorithms for RNNs are very challenging because internal recurrent paths form dependencies between two different time frames. In this paper, we first propose a generalized graph-based RNN structure that covers the most popular long short-term memory (LSTM) network. Then, we present a parallelization approach that automatically explores parallelisms of arbi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1503.02852","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":"2026-05-18T01:26:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Rvq+vaA5SKDfVLBw43NcUqWCPyjA3cumJb7dxFh/tkxeh3rsBHnhqm6GHOvfaO5ErA47b/KPsz+dfGBvKIoZAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-11T22:47:56.519109Z"},"content_sha256":"15f5aeac5377bb1865386a7d147056788387896449072a0cf0eba60041ba1c6e","schema_version":"1.0","event_id":"sha256:15f5aeac5377bb1865386a7d147056788387896449072a0cf0eba60041ba1c6e"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/75G6SZWEUWEIL74MBM376JGOXI/bundle.json","state_url":"https://pith.science/pith/75G6SZWEUWEIL74MBM376JGOXI/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/75G6SZWEUWEIL74MBM376JGOXI/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":"2026-06-11T22:47:56Z","links":{"resolver":"https://pith.science/pith/75G6SZWEUWEIL74MBM376JGOXI","bundle":"https://pith.science/pith/75G6SZWEUWEIL74MBM376JGOXI/bundle.json","state":"https://pith.science/pith/75G6SZWEUWEIL74MBM376JGOXI/state.json","well_known_bundle":"https://pith.science/.well-known/pith/75G6SZWEUWEIL74MBM376JGOXI/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2015:75G6SZWEUWEIL74MBM376JGOXI","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":"a6fe71ebcf47c43c74ce381d914b11b2462b6ae5727d613e76a3c80dbffd1fbf","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NE","submitted_at":"2015-03-10T10:27:55Z","title_canon_sha256":"10df6b001e9050eadd85820c536a12615d03547287ef9289e02d76c5c427e865"},"schema_version":"1.0","source":{"id":"1503.02852","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1503.02852","created_at":"2026-05-18T01:26:10Z"},{"alias_kind":"arxiv_version","alias_value":"1503.02852v1","created_at":"2026-05-18T01:26:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1503.02852","created_at":"2026-05-18T01:26:10Z"},{"alias_kind":"pith_short_12","alias_value":"75G6SZWEUWEI","created_at":"2026-05-18T12:29:07Z"},{"alias_kind":"pith_short_16","alias_value":"75G6SZWEUWEIL74M","created_at":"2026-05-18T12:29:07Z"},{"alias_kind":"pith_short_8","alias_value":"75G6SZWE","created_at":"2026-05-18T12:29:07Z"}],"graph_snapshots":[{"event_id":"sha256:15f5aeac5377bb1865386a7d147056788387896449072a0cf0eba60041ba1c6e","target":"graph","created_at":"2026-05-18T01:26:10Z","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) have shown outstanding performance on processing sequence data. However, they suffer from long training time, which demands parallel implementations of the training procedure. Parallelization of the training algorithms for RNNs are very challenging because internal recurrent paths form dependencies between two different time frames. In this paper, we first propose a generalized graph-based RNN structure that covers the most popular long short-term memory (LSTM) network. Then, we present a parallelization approach that automatically explores parallelisms of arbi","authors_text":"Kyuyeon Hwang, Wonyong Sung","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NE","submitted_at":"2015-03-10T10:27:55Z","title":"Single stream parallelization of generalized LSTM-like RNNs on a GPU"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1503.02852","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":"sha256:9f9d58ee8b60a2ba8f6fb69fd490758ad7c5e8aa80fe188646ffcb56bf48e6c2","target":"record","created_at":"2026-05-18T01:26:10Z","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":"a6fe71ebcf47c43c74ce381d914b11b2462b6ae5727d613e76a3c80dbffd1fbf","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NE","submitted_at":"2015-03-10T10:27:55Z","title_canon_sha256":"10df6b001e9050eadd85820c536a12615d03547287ef9289e02d76c5c427e865"},"schema_version":"1.0","source":{"id":"1503.02852","kind":"arxiv","version":1}},"canonical_sha256":"ff4de966c4a58885ff8c0b37ff24ceba27e0d0cb5bce65bf95a069c8b887fc1d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ff4de966c4a58885ff8c0b37ff24ceba27e0d0cb5bce65bf95a069c8b887fc1d","first_computed_at":"2026-05-18T01:26:10.543752Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T01:26:10.543752Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"TsJciua/VBHJRSsvxtMhKVisAlXgCIPdiAatw8htkDxQjgCV64MVTwwpBrJ0T49aFhuOhh+GVS73+2v1ImZeCQ==","signature_status":"signed_v1","signed_at":"2026-05-18T01:26:10.544354Z","signed_message":"canonical_sha256_bytes"},"source_id":"1503.02852","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9f9d58ee8b60a2ba8f6fb69fd490758ad7c5e8aa80fe188646ffcb56bf48e6c2","sha256:15f5aeac5377bb1865386a7d147056788387896449072a0cf0eba60041ba1c6e"],"state_sha256":"9d48657d109ccf3cef4f217ff850de689251ec7456689b751b14664d4635bce6"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"d+au1xM5hLNA4MOxe3J7c7M4tcHvsXi701xRq9PnNSrH32q+kvIPVpj4XHsdbyqkFWvrbN83NYYmzp4y5GybDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-11T22:47:56.523306Z","bundle_sha256":"719c507f92713c47663ff3fa7b2858070d5959102f854021665918d95362f320"}}