{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:HSEBSYRDAI6W7MUXYVVTIWURCR","short_pith_number":"pith:HSEBSYRD","schema_version":"1.0","canonical_sha256":"3c88196223023d6fb297c56b345a91146431b8f7bd2e265db2164f4ba0021280","source":{"kind":"arxiv","id":"1911.09483","version":1},"attestation_state":"computed","paper":{"title":"MUSE: Parallel Multi-Scale Attention for Sequence to Sequence Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Guangxiang Zhao, Jingjing Xu, Liangchen Luo, Xu Sun, Zhiyuan Zhang","submitted_at":"2019-11-17T09:36:07Z","abstract_excerpt":"In sequence to sequence learning, the self-attention mechanism proves to be highly effective, and achieves significant improvements in many tasks. However, the self-attention mechanism is not without its own flaws. Although self-attention can model extremely long dependencies, the attention in deep layers tends to overconcentrate on a single token, leading to insufficient use of local information and difficultly in representing long sequences. In this work, we explore parallel multi-scale representation learning on sequence data, striving to capture both long-range and short-range language str"},"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":"1911.09483","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-11-17T09:36:07Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"4a8862a54cff40f421b07b8f1735e726401d4732d86643ca5ca66077704c8672","abstract_canon_sha256":"513e74e36d34edbc07cc44c43adc4b9412da4e72f8407e3f1a8f6a705c29d0ee"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:21:00.840128Z","signature_b64":"znxW7mirhcah4aQp8oV7xE7KqWKszY41XeguUTi8uUkHfv/mznJLznwpM3VmkJfGEOtdi4K3lSP6ZXtxGm6UBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3c88196223023d6fb297c56b345a91146431b8f7bd2e265db2164f4ba0021280","last_reissued_at":"2026-07-05T00:21:00.839700Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:21:00.839700Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MUSE: Parallel Multi-Scale Attention for Sequence to Sequence Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Guangxiang Zhao, Jingjing Xu, Liangchen Luo, Xu Sun, Zhiyuan Zhang","submitted_at":"2019-11-17T09:36:07Z","abstract_excerpt":"In sequence to sequence learning, the self-attention mechanism proves to be highly effective, and achieves significant improvements in many tasks. However, the self-attention mechanism is not without its own flaws. Although self-attention can model extremely long dependencies, the attention in deep layers tends to overconcentrate on a single token, leading to insufficient use of local information and difficultly in representing long sequences. In this work, we explore parallel multi-scale representation learning on sequence data, striving to capture both long-range and short-range language str"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1911.09483","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/1911.09483/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":"1911.09483","created_at":"2026-07-05T00:21:00.839759+00:00"},{"alias_kind":"arxiv_version","alias_value":"1911.09483v1","created_at":"2026-07-05T00:21:00.839759+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1911.09483","created_at":"2026-07-05T00:21:00.839759+00:00"},{"alias_kind":"pith_short_12","alias_value":"HSEBSYRDAI6W","created_at":"2026-07-05T00:21:00.839759+00:00"},{"alias_kind":"pith_short_16","alias_value":"HSEBSYRDAI6W7MUX","created_at":"2026-07-05T00:21:00.839759+00:00"},{"alias_kind":"pith_short_8","alias_value":"HSEBSYRD","created_at":"2026-07-05T00:21:00.839759+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.01781","citing_title":"Structure-Guided Adaptive Propagation for Protein-Protein Interaction Site Prediction","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2510.23912","citing_title":"Key and Value Weights Are Probably All You Need: On the Necessity of the Query, Key, Value weight Triplet in Self-Attention Transformers","ref_index":24,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HSEBSYRDAI6W7MUXYVVTIWURCR","json":"https://pith.science/pith/HSEBSYRDAI6W7MUXYVVTIWURCR.json","graph_json":"https://pith.science/api/pith-number/HSEBSYRDAI6W7MUXYVVTIWURCR/graph.json","events_json":"https://pith.science/api/pith-number/HSEBSYRDAI6W7MUXYVVTIWURCR/events.json","paper":"https://pith.science/paper/HSEBSYRD"},"agent_actions":{"view_html":"https://pith.science/pith/HSEBSYRDAI6W7MUXYVVTIWURCR","download_json":"https://pith.science/pith/HSEBSYRDAI6W7MUXYVVTIWURCR.json","view_paper":"https://pith.science/paper/HSEBSYRD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1911.09483&json=true","fetch_graph":"https://pith.science/api/pith-number/HSEBSYRDAI6W7MUXYVVTIWURCR/graph.json","fetch_events":"https://pith.science/api/pith-number/HSEBSYRDAI6W7MUXYVVTIWURCR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HSEBSYRDAI6W7MUXYVVTIWURCR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HSEBSYRDAI6W7MUXYVVTIWURCR/action/storage_attestation","attest_author":"https://pith.science/pith/HSEBSYRDAI6W7MUXYVVTIWURCR/action/author_attestation","sign_citation":"https://pith.science/pith/HSEBSYRDAI6W7MUXYVVTIWURCR/action/citation_signature","submit_replication":"https://pith.science/pith/HSEBSYRDAI6W7MUXYVVTIWURCR/action/replication_record"}},"created_at":"2026-07-05T00:21:00.839759+00:00","updated_at":"2026-07-05T00:21:00.839759+00:00"}