{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:X5TS63XM6VM5PTHE7RPC7VJMPW","short_pith_number":"pith:X5TS63XM","schema_version":"1.0","canonical_sha256":"bf672f6eecf559d7cce4fc5e2fd52c7da2e1bcb8f5eb53c8fc95aa1e3b80e68e","source":{"kind":"arxiv","id":"2211.13491","version":1},"attestation_state":"computed","paper":{"title":"Spatial Mixture-of-Experts","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Nikoli Dryden, Torsten Hoefler","submitted_at":"2022-11-24T09:31:02Z","abstract_excerpt":"Many data have an underlying dependence on spatial location; it may be weather on the Earth, a simulation on a mesh, or a registered image. Yet this feature is rarely taken advantage of, and violates common assumptions made by many neural network layers, such as translation equivariance. Further, many works that do incorporate locality fail to capture fine-grained structure. To address this, we introduce the Spatial Mixture-of-Experts (SMoE) layer, a sparsely-gated layer that learns spatial structure in the input domain and routes experts at a fine-grained level to utilize it. We also develop "},"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":"2211.13491","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-11-24T09:31:02Z","cross_cats_sorted":[],"title_canon_sha256":"1f89c52e8f31bccaf8e07ef478cc33d4bbe6159f3b4f365badb2cfcce9341f3b","abstract_canon_sha256":"273aa482ba52cbe1c5e99eba74eaa561784d9a58e001e79331a0e7ba029dd7e8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:19:22.870717Z","signature_b64":"jHNhXokDSMzImqP2x5feZmzntDX2lk9s8up6Ylu2dDbYS5O6518JdXKGN0y5hAtpm93t+pTSh1FH22AeBSgOBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bf672f6eecf559d7cce4fc5e2fd52c7da2e1bcb8f5eb53c8fc95aa1e3b80e68e","last_reissued_at":"2026-07-05T05:19:22.870288Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:19:22.870288Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Spatial Mixture-of-Experts","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Nikoli Dryden, Torsten Hoefler","submitted_at":"2022-11-24T09:31:02Z","abstract_excerpt":"Many data have an underlying dependence on spatial location; it may be weather on the Earth, a simulation on a mesh, or a registered image. Yet this feature is rarely taken advantage of, and violates common assumptions made by many neural network layers, such as translation equivariance. Further, many works that do incorporate locality fail to capture fine-grained structure. To address this, we introduce the Spatial Mixture-of-Experts (SMoE) layer, a sparsely-gated layer that learns spatial structure in the input domain and routes experts at a fine-grained level to utilize it. We also develop "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2211.13491","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/2211.13491/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":"2211.13491","created_at":"2026-07-05T05:19:22.870356+00:00"},{"alias_kind":"arxiv_version","alias_value":"2211.13491v1","created_at":"2026-07-05T05:19:22.870356+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2211.13491","created_at":"2026-07-05T05:19:22.870356+00:00"},{"alias_kind":"pith_short_12","alias_value":"X5TS63XM6VM5","created_at":"2026-07-05T05:19:22.870356+00:00"},{"alias_kind":"pith_short_16","alias_value":"X5TS63XM6VM5PTHE","created_at":"2026-07-05T05:19:22.870356+00:00"},{"alias_kind":"pith_short_8","alias_value":"X5TS63XM","created_at":"2026-07-05T05:19:22.870356+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.22919","citing_title":"Hecto: Modular Sparse Experts for Adaptive and Interpretable Reasoning","ref_index":5,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/X5TS63XM6VM5PTHE7RPC7VJMPW","json":"https://pith.science/pith/X5TS63XM6VM5PTHE7RPC7VJMPW.json","graph_json":"https://pith.science/api/pith-number/X5TS63XM6VM5PTHE7RPC7VJMPW/graph.json","events_json":"https://pith.science/api/pith-number/X5TS63XM6VM5PTHE7RPC7VJMPW/events.json","paper":"https://pith.science/paper/X5TS63XM"},"agent_actions":{"view_html":"https://pith.science/pith/X5TS63XM6VM5PTHE7RPC7VJMPW","download_json":"https://pith.science/pith/X5TS63XM6VM5PTHE7RPC7VJMPW.json","view_paper":"https://pith.science/paper/X5TS63XM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2211.13491&json=true","fetch_graph":"https://pith.science/api/pith-number/X5TS63XM6VM5PTHE7RPC7VJMPW/graph.json","fetch_events":"https://pith.science/api/pith-number/X5TS63XM6VM5PTHE7RPC7VJMPW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/X5TS63XM6VM5PTHE7RPC7VJMPW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/X5TS63XM6VM5PTHE7RPC7VJMPW/action/storage_attestation","attest_author":"https://pith.science/pith/X5TS63XM6VM5PTHE7RPC7VJMPW/action/author_attestation","sign_citation":"https://pith.science/pith/X5TS63XM6VM5PTHE7RPC7VJMPW/action/citation_signature","submit_replication":"https://pith.science/pith/X5TS63XM6VM5PTHE7RPC7VJMPW/action/replication_record"}},"created_at":"2026-07-05T05:19:22.870356+00:00","updated_at":"2026-07-05T05:19:22.870356+00:00"}