{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:LJ5BFQ7H2E6EX4PSUDWY4EBUQD","short_pith_number":"pith:LJ5BFQ7H","schema_version":"1.0","canonical_sha256":"5a7a12c3e7d13c4bf1f2a0ed8e103480e97cae757fe11b1b0713298b8f658f07","source":{"kind":"arxiv","id":"2006.04439","version":4},"attestation_state":"computed","paper":{"title":"Liquid Time-constant Networks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.NE","stat.ML"],"primary_cat":"cs.LG","authors_text":"Alexander Amini, Daniela Rus, Mathias Lechner, Radu Grosu, Ramin Hasani","submitted_at":"2020-06-08T09:53:35Z","abstract_excerpt":"We introduce a new class of time-continuous recurrent neural network models. Instead of declaring a learning system's dynamics by implicit nonlinearities, we construct networks of linear first-order dynamical systems modulated via nonlinear interlinked gates. The resulting models represent dynamical systems with varying (i.e., liquid) time-constants coupled to their hidden state, with outputs being computed by numerical differential equation solvers. These neural networks exhibit stable and bounded behavior, yield superior expressivity within the family of neural ordinary differential equation"},"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":"2006.04439","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2020-06-08T09:53:35Z","cross_cats_sorted":["cs.NE","stat.ML"],"title_canon_sha256":"6f9541740432ad10544505b8cea551e60778db1c978e17f1bc699b752d09afd9","abstract_canon_sha256":"0b2f87468a9a40a0226b3aed504f7776272c944494e256f5dc6ccf6f10a77765"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:59:30.136349Z","signature_b64":"HnaeR27HQoKBD0q/Mn53EIg5YabIZJ+ioTGWFNTjMHYC+1XuS2yhvw8RjQrytnRmgkr61HT/3ttbFX9vgzV5BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5a7a12c3e7d13c4bf1f2a0ed8e103480e97cae757fe11b1b0713298b8f658f07","last_reissued_at":"2026-07-05T01:59:30.135785Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:59:30.135785Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Liquid Time-constant Networks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.NE","stat.ML"],"primary_cat":"cs.LG","authors_text":"Alexander Amini, Daniela Rus, Mathias Lechner, Radu Grosu, Ramin Hasani","submitted_at":"2020-06-08T09:53:35Z","abstract_excerpt":"We introduce a new class of time-continuous recurrent neural network models. Instead of declaring a learning system's dynamics by implicit nonlinearities, we construct networks of linear first-order dynamical systems modulated via nonlinear interlinked gates. The resulting models represent dynamical systems with varying (i.e., liquid) time-constants coupled to their hidden state, with outputs being computed by numerical differential equation solvers. These neural networks exhibit stable and bounded behavior, yield superior expressivity within the family of neural ordinary differential equation"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2006.04439","kind":"arxiv","version":4},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2006.04439/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"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":"2006.04439","created_at":"2026-07-05T01:59:30.135869+00:00"},{"alias_kind":"arxiv_version","alias_value":"2006.04439v4","created_at":"2026-07-05T01:59:30.135869+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2006.04439","created_at":"2026-07-05T01:59:30.135869+00:00"},{"alias_kind":"pith_short_12","alias_value":"LJ5BFQ7H2E6E","created_at":"2026-07-05T01:59:30.135869+00:00"},{"alias_kind":"pith_short_16","alias_value":"LJ5BFQ7H2E6EX4PS","created_at":"2026-07-05T01:59:30.135869+00:00"},{"alias_kind":"pith_short_8","alias_value":"LJ5BFQ7H","created_at":"2026-07-05T01:59:30.135869+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.07670","citing_title":"Liquid Neural Networks as a Drop-in Continuous-Time Deformation Field for Dynamic 3D Gaussian Splatting","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12874","citing_title":"LIFE -- an energy efficient advanced continual learning agentic AI framework for frontier systems","ref_index":19,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LJ5BFQ7H2E6EX4PSUDWY4EBUQD","json":"https://pith.science/pith/LJ5BFQ7H2E6EX4PSUDWY4EBUQD.json","graph_json":"https://pith.science/api/pith-number/LJ5BFQ7H2E6EX4PSUDWY4EBUQD/graph.json","events_json":"https://pith.science/api/pith-number/LJ5BFQ7H2E6EX4PSUDWY4EBUQD/events.json","paper":"https://pith.science/paper/LJ5BFQ7H"},"agent_actions":{"view_html":"https://pith.science/pith/LJ5BFQ7H2E6EX4PSUDWY4EBUQD","download_json":"https://pith.science/pith/LJ5BFQ7H2E6EX4PSUDWY4EBUQD.json","view_paper":"https://pith.science/paper/LJ5BFQ7H","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2006.04439&json=true","fetch_graph":"https://pith.science/api/pith-number/LJ5BFQ7H2E6EX4PSUDWY4EBUQD/graph.json","fetch_events":"https://pith.science/api/pith-number/LJ5BFQ7H2E6EX4PSUDWY4EBUQD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LJ5BFQ7H2E6EX4PSUDWY4EBUQD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LJ5BFQ7H2E6EX4PSUDWY4EBUQD/action/storage_attestation","attest_author":"https://pith.science/pith/LJ5BFQ7H2E6EX4PSUDWY4EBUQD/action/author_attestation","sign_citation":"https://pith.science/pith/LJ5BFQ7H2E6EX4PSUDWY4EBUQD/action/citation_signature","submit_replication":"https://pith.science/pith/LJ5BFQ7H2E6EX4PSUDWY4EBUQD/action/replication_record"}},"created_at":"2026-07-05T01:59:30.135869+00:00","updated_at":"2026-07-05T01:59:30.135869+00:00"}