{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:Q3OOSETV4VO53XQFGSAUGSH7Q6","short_pith_number":"pith:Q3OOSETV","schema_version":"1.0","canonical_sha256":"86dce91275e55dddde0534814348ff87a315668ecb5ce8a42d7032dffeb60ea3","source":{"kind":"arxiv","id":"2203.09176","version":1},"attestation_state":"computed","paper":{"title":"ODE Transformer: An Ordinary Differential Equation-Inspired Model for Sequence Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bei Li, Jingbo Zhu, Min Zhang, Quan Du, Shuhan Zhou, Tao Zhou, Tong Xiao, Xin Zeng, Xuebo Liu, Yi Jing","submitted_at":"2022-03-17T08:54:31Z","abstract_excerpt":"Residual networks are an Euler discretization of solutions to Ordinary Differential Equations (ODE). This paper explores a deeper relationship between Transformer and numerical ODE methods. We first show that a residual block of layers in Transformer can be described as a higher-order solution to ODE. Inspired by this, we design a new architecture, {\\it ODE Transformer}, which is analogous to the Runge-Kutta method that is well motivated in ODE. As a natural extension to Transformer, ODE Transformer is easy to implement and efficient to use. Experimental results on the large-scale machine tran"},"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":"2203.09176","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-03-17T08:54:31Z","cross_cats_sorted":[],"title_canon_sha256":"70be76def8a937a217e2472bf3e20fd462da99b626f284d23df002053cf3a7f8","abstract_canon_sha256":"2712ca558f9cd90073dfba3a18d89a0efd3e9a42e71fb9946aa96dd17e786e05"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:13:54.580081Z","signature_b64":"jYe8fsOJJrft5GPo70gQlt5dGURUhjb4HmxaLgGJ81YYGKa70nYi52cxxGzea5bdY4WwoUZmXlJag9dYhcqCDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"86dce91275e55dddde0534814348ff87a315668ecb5ce8a42d7032dffeb60ea3","last_reissued_at":"2026-07-05T04:13:54.579550Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:13:54.579550Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ODE Transformer: An Ordinary Differential Equation-Inspired Model for Sequence Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bei Li, Jingbo Zhu, Min Zhang, Quan Du, Shuhan Zhou, Tao Zhou, Tong Xiao, Xin Zeng, Xuebo Liu, Yi Jing","submitted_at":"2022-03-17T08:54:31Z","abstract_excerpt":"Residual networks are an Euler discretization of solutions to Ordinary Differential Equations (ODE). This paper explores a deeper relationship between Transformer and numerical ODE methods. We first show that a residual block of layers in Transformer can be described as a higher-order solution to ODE. Inspired by this, we design a new architecture, {\\it ODE Transformer}, which is analogous to the Runge-Kutta method that is well motivated in ODE. As a natural extension to Transformer, ODE Transformer is easy to implement and efficient to use. Experimental results on the large-scale machine tran"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.09176","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2203.09176/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":"2203.09176","created_at":"2026-07-05T04:13:54.579607+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.09176v1","created_at":"2026-07-05T04:13:54.579607+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.09176","created_at":"2026-07-05T04:13:54.579607+00:00"},{"alias_kind":"pith_short_12","alias_value":"Q3OOSETV4VO5","created_at":"2026-07-05T04:13:54.579607+00:00"},{"alias_kind":"pith_short_16","alias_value":"Q3OOSETV4VO53XQF","created_at":"2026-07-05T04:13:54.579607+00:00"},{"alias_kind":"pith_short_8","alias_value":"Q3OOSETV","created_at":"2026-07-05T04:13:54.579607+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2504.18590","citing_title":"A multilevel approach to accelerate the training of Transformers","ref_index":18,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Q3OOSETV4VO53XQFGSAUGSH7Q6","json":"https://pith.science/pith/Q3OOSETV4VO53XQFGSAUGSH7Q6.json","graph_json":"https://pith.science/api/pith-number/Q3OOSETV4VO53XQFGSAUGSH7Q6/graph.json","events_json":"https://pith.science/api/pith-number/Q3OOSETV4VO53XQFGSAUGSH7Q6/events.json","paper":"https://pith.science/paper/Q3OOSETV"},"agent_actions":{"view_html":"https://pith.science/pith/Q3OOSETV4VO53XQFGSAUGSH7Q6","download_json":"https://pith.science/pith/Q3OOSETV4VO53XQFGSAUGSH7Q6.json","view_paper":"https://pith.science/paper/Q3OOSETV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.09176&json=true","fetch_graph":"https://pith.science/api/pith-number/Q3OOSETV4VO53XQFGSAUGSH7Q6/graph.json","fetch_events":"https://pith.science/api/pith-number/Q3OOSETV4VO53XQFGSAUGSH7Q6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Q3OOSETV4VO53XQFGSAUGSH7Q6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Q3OOSETV4VO53XQFGSAUGSH7Q6/action/storage_attestation","attest_author":"https://pith.science/pith/Q3OOSETV4VO53XQFGSAUGSH7Q6/action/author_attestation","sign_citation":"https://pith.science/pith/Q3OOSETV4VO53XQFGSAUGSH7Q6/action/citation_signature","submit_replication":"https://pith.science/pith/Q3OOSETV4VO53XQFGSAUGSH7Q6/action/replication_record"}},"created_at":"2026-07-05T04:13:54.579607+00:00","updated_at":"2026-07-05T04:13:54.579607+00:00"}