{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:RS5SBHLTSHW6PI772FFRNRU4EZ","short_pith_number":"pith:RS5SBHLT","schema_version":"1.0","canonical_sha256":"8cbb209d7391ede7a3ffd14b16c69c266412438d8b3a972a8d434ef533eeed0a","source":{"kind":"arxiv","id":"2310.14921","version":2},"attestation_state":"computed","paper":{"title":"PartialFormer: Modeling Part Instead of Whole for Machine Translation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bei Li, Huiwen Bao, Jiale Wang, Jingbo Zhu, Tong Xiao, Tong Zheng, Weiqiao Shan","submitted_at":"2023-10-23T13:25:54Z","abstract_excerpt":"The design choices in Transformer feed-forward neural networks have resulted in significant computational and parameter overhead. In this work, we emphasize the importance of hidden dimensions in designing lightweight FFNs, a factor often overlooked in previous architectures. Guided by this principle, we introduce PartialFormer, a parameter-efficient Transformer architecture utilizing multiple smaller FFNs to reduce parameters and computation while maintaining essential hidden dimensions. These smaller FFNs are integrated into a multi-head attention mechanism for effective collaboration. We al"},"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":"2310.14921","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-10-23T13:25:54Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"73793bde0a7d17d2207c99249b318129671f72be09c676eb4e058cb1a188e643","abstract_canon_sha256":"f282da9a492137e5c4d925528f6fb70601a4ba6d9735f12b4f23de2462d327db"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:27:43.349897Z","signature_b64":"XHTFGypw56tXTE4UZbdisiH21HpNRi6Rbr8pIhHgW8cQr357YtMEsDKUrlGNaM9mdUtkqgHu9tWQHbronjR9Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8cbb209d7391ede7a3ffd14b16c69c266412438d8b3a972a8d434ef533eeed0a","last_reissued_at":"2026-07-05T08:27:43.349349Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:27:43.349349Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PartialFormer: Modeling Part Instead of Whole for Machine Translation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bei Li, Huiwen Bao, Jiale Wang, Jingbo Zhu, Tong Xiao, Tong Zheng, Weiqiao Shan","submitted_at":"2023-10-23T13:25:54Z","abstract_excerpt":"The design choices in Transformer feed-forward neural networks have resulted in significant computational and parameter overhead. In this work, we emphasize the importance of hidden dimensions in designing lightweight FFNs, a factor often overlooked in previous architectures. Guided by this principle, we introduce PartialFormer, a parameter-efficient Transformer architecture utilizing multiple smaller FFNs to reduce parameters and computation while maintaining essential hidden dimensions. These smaller FFNs are integrated into a multi-head attention mechanism for effective collaboration. We al"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.14921","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2310.14921/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":"2310.14921","created_at":"2026-07-05T08:27:43.349409+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.14921v2","created_at":"2026-07-05T08:27:43.349409+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.14921","created_at":"2026-07-05T08:27:43.349409+00:00"},{"alias_kind":"pith_short_12","alias_value":"RS5SBHLTSHW6","created_at":"2026-07-05T08:27:43.349409+00:00"},{"alias_kind":"pith_short_16","alias_value":"RS5SBHLTSHW6PI77","created_at":"2026-07-05T08:27:43.349409+00:00"},{"alias_kind":"pith_short_8","alias_value":"RS5SBHLT","created_at":"2026-07-05T08:27:43.349409+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/RS5SBHLTSHW6PI772FFRNRU4EZ","json":"https://pith.science/pith/RS5SBHLTSHW6PI772FFRNRU4EZ.json","graph_json":"https://pith.science/api/pith-number/RS5SBHLTSHW6PI772FFRNRU4EZ/graph.json","events_json":"https://pith.science/api/pith-number/RS5SBHLTSHW6PI772FFRNRU4EZ/events.json","paper":"https://pith.science/paper/RS5SBHLT"},"agent_actions":{"view_html":"https://pith.science/pith/RS5SBHLTSHW6PI772FFRNRU4EZ","download_json":"https://pith.science/pith/RS5SBHLTSHW6PI772FFRNRU4EZ.json","view_paper":"https://pith.science/paper/RS5SBHLT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.14921&json=true","fetch_graph":"https://pith.science/api/pith-number/RS5SBHLTSHW6PI772FFRNRU4EZ/graph.json","fetch_events":"https://pith.science/api/pith-number/RS5SBHLTSHW6PI772FFRNRU4EZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RS5SBHLTSHW6PI772FFRNRU4EZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RS5SBHLTSHW6PI772FFRNRU4EZ/action/storage_attestation","attest_author":"https://pith.science/pith/RS5SBHLTSHW6PI772FFRNRU4EZ/action/author_attestation","sign_citation":"https://pith.science/pith/RS5SBHLTSHW6PI772FFRNRU4EZ/action/citation_signature","submit_replication":"https://pith.science/pith/RS5SBHLTSHW6PI772FFRNRU4EZ/action/replication_record"}},"created_at":"2026-07-05T08:27:43.349409+00:00","updated_at":"2026-07-05T08:27:43.349409+00:00"}