{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:PKRDQIMDISY7WHNNZR7H32UGJQ","short_pith_number":"pith:PKRDQIMD","schema_version":"1.0","canonical_sha256":"7aa238218344b1fb1dadcc7e7dea864c079089eb7e963e7d1b67f75b68b20cf2","source":{"kind":"arxiv","id":"1902.09113","version":3},"attestation_state":"computed","paper":{"title":"Star-Transformer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Pengfei Liu, Qipeng Guo, Xiangyang Xue, Xipeng Qiu, Yunfan Shao, Zheng Zhang","submitted_at":"2019-02-25T07:07:38Z","abstract_excerpt":"Although Transformer has achieved great successes on many NLP tasks, its heavy structure with fully-connected attention connections leads to dependencies on large training data. In this paper, we present Star-Transformer, a lightweight alternative by careful sparsification. To reduce model complexity, we replace the fully-connected structure with a star-shaped topology, in which every two non-adjacent nodes are connected through a shared relay node. Thus, complexity is reduced from quadratic to linear, while preserving capacity to capture both local composition and long-range dependency. The e"},"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":"1902.09113","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-02-25T07:07:38Z","cross_cats_sorted":[],"title_canon_sha256":"41b9fcd2258706575c555d95ca08e294c1ca16d2ba24ecc5e0b349d867d7ea4a","abstract_canon_sha256":"c77d32fa69100ca4da7037e5f1b5229e708b63dc1588663c78523474486faeb9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:17:11.474960Z","signature_b64":"WJk92Uy+4k5kaAQLNtG6WcNppalLWFCCVl2ovuqrXbXVPOyaqIzSyZ+L9fmldFpd9Jd8tVjooR5lAUaTO3jGCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7aa238218344b1fb1dadcc7e7dea864c079089eb7e963e7d1b67f75b68b20cf2","last_reissued_at":"2026-07-05T04:17:11.474454Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:17:11.474454Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Star-Transformer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Pengfei Liu, Qipeng Guo, Xiangyang Xue, Xipeng Qiu, Yunfan Shao, Zheng Zhang","submitted_at":"2019-02-25T07:07:38Z","abstract_excerpt":"Although Transformer has achieved great successes on many NLP tasks, its heavy structure with fully-connected attention connections leads to dependencies on large training data. In this paper, we present Star-Transformer, a lightweight alternative by careful sparsification. To reduce model complexity, we replace the fully-connected structure with a star-shaped topology, in which every two non-adjacent nodes are connected through a shared relay node. Thus, complexity is reduced from quadratic to linear, while preserving capacity to capture both local composition and long-range dependency. The e"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1902.09113","kind":"arxiv","version":3},"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/1902.09113/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":"1902.09113","created_at":"2026-07-05T04:17:11.474517+00:00"},{"alias_kind":"arxiv_version","alias_value":"1902.09113v3","created_at":"2026-07-05T04:17:11.474517+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1902.09113","created_at":"2026-07-05T04:17:11.474517+00:00"},{"alias_kind":"pith_short_12","alias_value":"PKRDQIMDISY7","created_at":"2026-07-05T04:17:11.474517+00:00"},{"alias_kind":"pith_short_16","alias_value":"PKRDQIMDISY7WHNN","created_at":"2026-07-05T04:17:11.474517+00:00"},{"alias_kind":"pith_short_8","alias_value":"PKRDQIMD","created_at":"2026-07-05T04:17:11.474517+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":8,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"1906.12330","citing_title":"Graph Star Net for Generalized Multi-Task Learning","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2512.13368","citing_title":"BlossomRec: Block-level Fused Sparse Attention Mechanism for Sequential Recommendations","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"1907.06607","citing_title":"Agglomerative Attention","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2308.08089","citing_title":"DragNUWA: Fine-grained Control in Video Generation by Integrating Text, Image, and Trajectory","ref_index":275,"is_internal_anchor":false},{"citing_arxiv_id":"2512.12602","citing_title":"Exact Flow Linear Attention: Exact Solution from Continuous-Time Dynamics","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2502.13189","citing_title":"MoBA: Mixture of Block Attention for Long-Context LLMs","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2510.26692","citing_title":"Kimi Linear: An Expressive, Efficient Attention Architecture","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18939","citing_title":"TabEmb: Joint Semantic-Structure Embedding for Table Annotation","ref_index":92,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PKRDQIMDISY7WHNNZR7H32UGJQ","json":"https://pith.science/pith/PKRDQIMDISY7WHNNZR7H32UGJQ.json","graph_json":"https://pith.science/api/pith-number/PKRDQIMDISY7WHNNZR7H32UGJQ/graph.json","events_json":"https://pith.science/api/pith-number/PKRDQIMDISY7WHNNZR7H32UGJQ/events.json","paper":"https://pith.science/paper/PKRDQIMD"},"agent_actions":{"view_html":"https://pith.science/pith/PKRDQIMDISY7WHNNZR7H32UGJQ","download_json":"https://pith.science/pith/PKRDQIMDISY7WHNNZR7H32UGJQ.json","view_paper":"https://pith.science/paper/PKRDQIMD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1902.09113&json=true","fetch_graph":"https://pith.science/api/pith-number/PKRDQIMDISY7WHNNZR7H32UGJQ/graph.json","fetch_events":"https://pith.science/api/pith-number/PKRDQIMDISY7WHNNZR7H32UGJQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PKRDQIMDISY7WHNNZR7H32UGJQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PKRDQIMDISY7WHNNZR7H32UGJQ/action/storage_attestation","attest_author":"https://pith.science/pith/PKRDQIMDISY7WHNNZR7H32UGJQ/action/author_attestation","sign_citation":"https://pith.science/pith/PKRDQIMDISY7WHNNZR7H32UGJQ/action/citation_signature","submit_replication":"https://pith.science/pith/PKRDQIMDISY7WHNNZR7H32UGJQ/action/replication_record"}},"created_at":"2026-07-05T04:17:11.474517+00:00","updated_at":"2026-07-05T04:17:11.474517+00:00"}