{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:OMZXCKRVXT4PLTMAMMNGX55F3E","short_pith_number":"pith:OMZXCKRV","schema_version":"1.0","canonical_sha256":"7333712a35bcf8f5cd80631a6bf7a5d933aea757d42e3c9c271f08e58c87f890","source":{"kind":"arxiv","id":"1902.06704","version":1},"attestation_state":"computed","paper":{"title":"Towards Non-saturating Recurrent Units for Modelling Long-term Dependencies","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"cs.NE","authors_text":"Chinnadhurai Sankar, Eugene Vorontsov, Samira Ebrahimi Kahou, Sarath Chandar, Yoshua Bengio","submitted_at":"2019-01-22T15:24:27Z","abstract_excerpt":"Modelling long-term dependencies is a challenge for recurrent neural networks. This is primarily due to the fact that gradients vanish during training, as the sequence length increases. Gradients can be attenuated by transition operators and are attenuated or dropped by activation functions. Canonical architectures like LSTM alleviate this issue by skipping information through a memory mechanism. We propose a new recurrent architecture (Non-saturating Recurrent Unit; NRU) that relies on a memory mechanism but forgoes both saturating activation functions and saturating gates, in order to furthe"},"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":"1902.06704","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NE","submitted_at":"2019-01-22T15:24:27Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"fd07bbd05d41c336f04f17b166109e17c943390c7dfda71d38b9db8d345ab1f3","abstract_canon_sha256":"3a687ce2db33c55ce982026cd7800221909a395079f32d0aec8183e7a3f36885"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:53:44.746090Z","signature_b64":"Kmi0d0snShD4kHZKHdnwsJtgPQeLlH0VET+Fn+7Q0ARVeXiCxQZ8LEqkPEXiNsduvWrgxcYl+mjzGVYj2VCrCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7333712a35bcf8f5cd80631a6bf7a5d933aea757d42e3c9c271f08e58c87f890","last_reissued_at":"2026-05-17T23:53:44.745614Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:53:44.745614Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Towards Non-saturating Recurrent Units for Modelling Long-term Dependencies","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"cs.NE","authors_text":"Chinnadhurai Sankar, Eugene Vorontsov, Samira Ebrahimi Kahou, Sarath Chandar, Yoshua Bengio","submitted_at":"2019-01-22T15:24:27Z","abstract_excerpt":"Modelling long-term dependencies is a challenge for recurrent neural networks. This is primarily due to the fact that gradients vanish during training, as the sequence length increases. Gradients can be attenuated by transition operators and are attenuated or dropped by activation functions. Canonical architectures like LSTM alleviate this issue by skipping information through a memory mechanism. We propose a new recurrent architecture (Non-saturating Recurrent Unit; NRU) that relies on a memory mechanism but forgoes both saturating activation functions and saturating gates, in order to furthe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1902.06704","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"},"aliases":[{"alias_kind":"arxiv","alias_value":"1902.06704","created_at":"2026-05-17T23:53:44.745690+00:00"},{"alias_kind":"arxiv_version","alias_value":"1902.06704v1","created_at":"2026-05-17T23:53:44.745690+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1902.06704","created_at":"2026-05-17T23:53:44.745690+00:00"},{"alias_kind":"pith_short_12","alias_value":"OMZXCKRVXT4P","created_at":"2026-05-18T12:33:24.271573+00:00"},{"alias_kind":"pith_short_16","alias_value":"OMZXCKRVXT4PLTMA","created_at":"2026-05-18T12:33:24.271573+00:00"},{"alias_kind":"pith_short_8","alias_value":"OMZXCKRV","created_at":"2026-05-18T12:33:24.271573+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/OMZXCKRVXT4PLTMAMMNGX55F3E","json":"https://pith.science/pith/OMZXCKRVXT4PLTMAMMNGX55F3E.json","graph_json":"https://pith.science/api/pith-number/OMZXCKRVXT4PLTMAMMNGX55F3E/graph.json","events_json":"https://pith.science/api/pith-number/OMZXCKRVXT4PLTMAMMNGX55F3E/events.json","paper":"https://pith.science/paper/OMZXCKRV"},"agent_actions":{"view_html":"https://pith.science/pith/OMZXCKRVXT4PLTMAMMNGX55F3E","download_json":"https://pith.science/pith/OMZXCKRVXT4PLTMAMMNGX55F3E.json","view_paper":"https://pith.science/paper/OMZXCKRV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1902.06704&json=true","fetch_graph":"https://pith.science/api/pith-number/OMZXCKRVXT4PLTMAMMNGX55F3E/graph.json","fetch_events":"https://pith.science/api/pith-number/OMZXCKRVXT4PLTMAMMNGX55F3E/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OMZXCKRVXT4PLTMAMMNGX55F3E/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OMZXCKRVXT4PLTMAMMNGX55F3E/action/storage_attestation","attest_author":"https://pith.science/pith/OMZXCKRVXT4PLTMAMMNGX55F3E/action/author_attestation","sign_citation":"https://pith.science/pith/OMZXCKRVXT4PLTMAMMNGX55F3E/action/citation_signature","submit_replication":"https://pith.science/pith/OMZXCKRVXT4PLTMAMMNGX55F3E/action/replication_record"}},"created_at":"2026-05-17T23:53:44.745690+00:00","updated_at":"2026-05-17T23:53:44.745690+00:00"}