{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:JOBLC4UFTYAXO2ITECPTPNA5YF","short_pith_number":"pith:JOBLC4UF","schema_version":"1.0","canonical_sha256":"4b82b172859e01776913209f37b41dc14b7525ca7cce80e2795e588a3bdfc66c","source":{"kind":"arxiv","id":"2210.11794","version":2},"attestation_state":"computed","paper":{"title":"Diffuser: Efficient Transformers with Multi-hop Attention Diffusion for Long Sequences","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Aosong Feng, Irene Li, Rex Ying, Yuang Jiang","submitted_at":"2022-10-21T08:13:34Z","abstract_excerpt":"Efficient Transformers have been developed for long sequence modeling, due to their subquadratic memory and time complexity. Sparse Transformer is a popular approach to improving the efficiency of Transformers by restricting self-attention to locations specified by the predefined sparse patterns. However, leveraging sparsity may sacrifice expressiveness compared to full-attention, when important token correlations are multiple hops away. To combine advantages of both the efficiency of sparse transformer and the expressiveness of full-attention Transformer, we propose \\textit{Diffuser}, a new s"},"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":"2210.11794","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-10-21T08:13:34Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"c9a5718d3c9f87be15c54da91194dced231cee552c0d3be2ce8ade0bfd6ca9a1","abstract_canon_sha256":"0149068e17de92c4835b2057894f42575287ff923da70435b4db438bbfc38723"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:37:09.694067Z","signature_b64":"obnjxJocO7xzllbpYIGKGYpFwwu4fv5PSFrWaGktt/y33Xclns68dTJCN4qAJzDb+M5h0ZEYWKInygcg+7ZyCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4b82b172859e01776913209f37b41dc14b7525ca7cce80e2795e588a3bdfc66c","last_reissued_at":"2026-07-05T05:37:09.693582Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:37:09.693582Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Diffuser: Efficient Transformers with Multi-hop Attention Diffusion for Long Sequences","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Aosong Feng, Irene Li, Rex Ying, Yuang Jiang","submitted_at":"2022-10-21T08:13:34Z","abstract_excerpt":"Efficient Transformers have been developed for long sequence modeling, due to their subquadratic memory and time complexity. Sparse Transformer is a popular approach to improving the efficiency of Transformers by restricting self-attention to locations specified by the predefined sparse patterns. However, leveraging sparsity may sacrifice expressiveness compared to full-attention, when important token correlations are multiple hops away. To combine advantages of both the efficiency of sparse transformer and the expressiveness of full-attention Transformer, we propose \\textit{Diffuser}, a new s"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.11794","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/2210.11794/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":"2210.11794","created_at":"2026-07-05T05:37:09.693644+00:00"},{"alias_kind":"arxiv_version","alias_value":"2210.11794v2","created_at":"2026-07-05T05:37:09.693644+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.11794","created_at":"2026-07-05T05:37:09.693644+00:00"},{"alias_kind":"pith_short_12","alias_value":"JOBLC4UFTYAX","created_at":"2026-07-05T05:37:09.693644+00:00"},{"alias_kind":"pith_short_16","alias_value":"JOBLC4UFTYAXO2IT","created_at":"2026-07-05T05:37:09.693644+00:00"},{"alias_kind":"pith_short_8","alias_value":"JOBLC4UF","created_at":"2026-07-05T05:37:09.693644+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/JOBLC4UFTYAXO2ITECPTPNA5YF","json":"https://pith.science/pith/JOBLC4UFTYAXO2ITECPTPNA5YF.json","graph_json":"https://pith.science/api/pith-number/JOBLC4UFTYAXO2ITECPTPNA5YF/graph.json","events_json":"https://pith.science/api/pith-number/JOBLC4UFTYAXO2ITECPTPNA5YF/events.json","paper":"https://pith.science/paper/JOBLC4UF"},"agent_actions":{"view_html":"https://pith.science/pith/JOBLC4UFTYAXO2ITECPTPNA5YF","download_json":"https://pith.science/pith/JOBLC4UFTYAXO2ITECPTPNA5YF.json","view_paper":"https://pith.science/paper/JOBLC4UF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2210.11794&json=true","fetch_graph":"https://pith.science/api/pith-number/JOBLC4UFTYAXO2ITECPTPNA5YF/graph.json","fetch_events":"https://pith.science/api/pith-number/JOBLC4UFTYAXO2ITECPTPNA5YF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JOBLC4UFTYAXO2ITECPTPNA5YF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JOBLC4UFTYAXO2ITECPTPNA5YF/action/storage_attestation","attest_author":"https://pith.science/pith/JOBLC4UFTYAXO2ITECPTPNA5YF/action/author_attestation","sign_citation":"https://pith.science/pith/JOBLC4UFTYAXO2ITECPTPNA5YF/action/citation_signature","submit_replication":"https://pith.science/pith/JOBLC4UFTYAXO2ITECPTPNA5YF/action/replication_record"}},"created_at":"2026-07-05T05:37:09.693644+00:00","updated_at":"2026-07-05T05:37:09.693644+00:00"}