{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:KFMFPL27XUAEFEJUDIUZFSW2AS","short_pith_number":"pith:KFMFPL27","canonical_record":{"source":{"id":"1810.10927","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-10-25T15:27:23Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"9cdc1d564399fe0300de9c453c307fa5d670f813e7cb0b1a25d49ce0ed2d102c","abstract_canon_sha256":"c861a6df8cefbc75e27ec3f990e09e4590ca5dedfc6ece6282c628d7ec431f21"},"schema_version":"1.0"},"canonical_sha256":"515857af5fbd004291341a2992cada048169b5f588fa8a1c54cbe057e31e12af","source":{"kind":"arxiv","id":"1810.10927","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1810.10927","created_at":"2026-05-17T23:58:30Z"},{"alias_kind":"arxiv_version","alias_value":"1810.10927v2","created_at":"2026-05-17T23:58:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1810.10927","created_at":"2026-05-17T23:58:30Z"},{"alias_kind":"pith_short_12","alias_value":"KFMFPL27XUAE","created_at":"2026-05-18T12:32:33Z"},{"alias_kind":"pith_short_16","alias_value":"KFMFPL27XUAEFEJU","created_at":"2026-05-18T12:32:33Z"},{"alias_kind":"pith_short_8","alias_value":"KFMFPL27","created_at":"2026-05-18T12:32:33Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:KFMFPL27XUAEFEJUDIUZFSW2AS","target":"record","payload":{"canonical_record":{"source":{"id":"1810.10927","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-10-25T15:27:23Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"9cdc1d564399fe0300de9c453c307fa5d670f813e7cb0b1a25d49ce0ed2d102c","abstract_canon_sha256":"c861a6df8cefbc75e27ec3f990e09e4590ca5dedfc6ece6282c628d7ec431f21"},"schema_version":"1.0"},"canonical_sha256":"515857af5fbd004291341a2992cada048169b5f588fa8a1c54cbe057e31e12af","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:58:30.287558Z","signature_b64":"taxC6RxjhiuYklVagZ1m70gXXfa6Vr31R6beYskY+TJN3k8EsIq1x07S0M40BdUCupt9mQh35thBAnDj5VbODA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"515857af5fbd004291341a2992cada048169b5f588fa8a1c54cbe057e31e12af","last_reissued_at":"2026-05-17T23:58:30.286900Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:58:30.286900Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1810.10927","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:58:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9UE78kUW1odebD3uY6ETXMmQpxRS1oZJFAXj6MsV1opnj3ku5Ea/U0qGA/xHmcaq/Tdhmy4BxecPmYzgzTlTCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-19T16:54:04.747839Z"},"content_sha256":"5b8127c52f6958c0efa82ef99882801144513041597ffa1d02c7e0fb4d2d41ac","schema_version":"1.0","event_id":"sha256:5b8127c52f6958c0efa82ef99882801144513041597ffa1d02c7e0fb4d2d41ac"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:KFMFPL27XUAEFEJUDIUZFSW2AS","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Bayesian Compression for Natural Language Processing","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"cs.CL","authors_text":"Dmitry Vetrov, Ekaterina Lobacheva, Nadezhda Chirkova","submitted_at":"2018-10-25T15:27:23Z","abstract_excerpt":"In natural language processing, a lot of the tasks are successfully solved with recurrent neural networks, but such models have a huge number of parameters. The majority of these parameters are often concentrated in the embedding layer, which size grows proportionally to the vocabulary length. We propose a Bayesian sparsification technique for RNNs which allows compressing the RNN dozens or hundreds of times without time-consuming hyperparameters tuning. We also generalize the model for vocabulary sparsification to filter out unnecessary words and compress the RNN even further. We show that th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1810.10927","kind":"arxiv","version":2},"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":"2026-05-17T23:58:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Odd58nZN1Bl+EKi9Sr4bSEuOTrfqc/Kht+uogM1C4A4ZNx+0Vs0oZxTgu2z8Wz33X7j1LooriCx5KJLqP3fKAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-19T16:54:04.748586Z"},"content_sha256":"7575eac81ba31de3635903598a213418c524a71707f3bc6d7edb97b33c9936ee","schema_version":"1.0","event_id":"sha256:7575eac81ba31de3635903598a213418c524a71707f3bc6d7edb97b33c9936ee"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KFMFPL27XUAEFEJUDIUZFSW2AS/bundle.json","state_url":"https://pith.science/pith/KFMFPL27XUAEFEJUDIUZFSW2AS/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KFMFPL27XUAEFEJUDIUZFSW2AS/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":"2026-05-19T16:54:04Z","links":{"resolver":"https://pith.science/pith/KFMFPL27XUAEFEJUDIUZFSW2AS","bundle":"https://pith.science/pith/KFMFPL27XUAEFEJUDIUZFSW2AS/bundle.json","state":"https://pith.science/pith/KFMFPL27XUAEFEJUDIUZFSW2AS/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KFMFPL27XUAEFEJUDIUZFSW2AS/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:KFMFPL27XUAEFEJUDIUZFSW2AS","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":"c861a6df8cefbc75e27ec3f990e09e4590ca5dedfc6ece6282c628d7ec431f21","cross_cats_sorted":["cs.LG","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-10-25T15:27:23Z","title_canon_sha256":"9cdc1d564399fe0300de9c453c307fa5d670f813e7cb0b1a25d49ce0ed2d102c"},"schema_version":"1.0","source":{"id":"1810.10927","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1810.10927","created_at":"2026-05-17T23:58:30Z"},{"alias_kind":"arxiv_version","alias_value":"1810.10927v2","created_at":"2026-05-17T23:58:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1810.10927","created_at":"2026-05-17T23:58:30Z"},{"alias_kind":"pith_short_12","alias_value":"KFMFPL27XUAE","created_at":"2026-05-18T12:32:33Z"},{"alias_kind":"pith_short_16","alias_value":"KFMFPL27XUAEFEJU","created_at":"2026-05-18T12:32:33Z"},{"alias_kind":"pith_short_8","alias_value":"KFMFPL27","created_at":"2026-05-18T12:32:33Z"}],"graph_snapshots":[{"event_id":"sha256:7575eac81ba31de3635903598a213418c524a71707f3bc6d7edb97b33c9936ee","target":"graph","created_at":"2026-05-17T23:58:30Z","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":"In natural language processing, a lot of the tasks are successfully solved with recurrent neural networks, but such models have a huge number of parameters. The majority of these parameters are often concentrated in the embedding layer, which size grows proportionally to the vocabulary length. We propose a Bayesian sparsification technique for RNNs which allows compressing the RNN dozens or hundreds of times without time-consuming hyperparameters tuning. We also generalize the model for vocabulary sparsification to filter out unnecessary words and compress the RNN even further. We show that th","authors_text":"Dmitry Vetrov, Ekaterina Lobacheva, Nadezhda Chirkova","cross_cats":["cs.LG","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-10-25T15:27:23Z","title":"Bayesian Compression for Natural Language Processing"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1810.10927","kind":"arxiv","version":2},"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":"sha256:5b8127c52f6958c0efa82ef99882801144513041597ffa1d02c7e0fb4d2d41ac","target":"record","created_at":"2026-05-17T23:58:30Z","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":"c861a6df8cefbc75e27ec3f990e09e4590ca5dedfc6ece6282c628d7ec431f21","cross_cats_sorted":["cs.LG","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-10-25T15:27:23Z","title_canon_sha256":"9cdc1d564399fe0300de9c453c307fa5d670f813e7cb0b1a25d49ce0ed2d102c"},"schema_version":"1.0","source":{"id":"1810.10927","kind":"arxiv","version":2}},"canonical_sha256":"515857af5fbd004291341a2992cada048169b5f588fa8a1c54cbe057e31e12af","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"515857af5fbd004291341a2992cada048169b5f588fa8a1c54cbe057e31e12af","first_computed_at":"2026-05-17T23:58:30.286900Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:58:30.286900Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"taxC6RxjhiuYklVagZ1m70gXXfa6Vr31R6beYskY+TJN3k8EsIq1x07S0M40BdUCupt9mQh35thBAnDj5VbODA==","signature_status":"signed_v1","signed_at":"2026-05-17T23:58:30.287558Z","signed_message":"canonical_sha256_bytes"},"source_id":"1810.10927","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5b8127c52f6958c0efa82ef99882801144513041597ffa1d02c7e0fb4d2d41ac","sha256:7575eac81ba31de3635903598a213418c524a71707f3bc6d7edb97b33c9936ee"],"state_sha256":"ef6f80869d018a62062edf358484afabf85846121ba736865c2b47d33b9a7ec2"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"66O/W1Y728TCm9vJ6CjNcxz2Inj7cPkhccN4vBoGxHo9e794pOxMoffQxDlJi+f5fSe55EtRJ/+Sw6PRIqvPCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-19T16:54:04.752507Z","bundle_sha256":"bf16da533d76ce9c726ed53134541c4387eec1b64ad8b5b740f1fba21b3f24db"}}