{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:63LFR3DSWOBRZMAG3A46WNHIAA","short_pith_number":"pith:63LFR3DS","schema_version":"1.0","canonical_sha256":"f6d658ec72b3831cb006d839eb34e8002c8bfa0c98da76c29372ba5645a5f818","source":{"kind":"arxiv","id":"2402.12550","version":4},"attestation_state":"computed","paper":{"title":"Multilinear Mixture of Experts: Scalable Expert Specialization through Factorization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Christos Tzelepis, Grigorios G. Chrysos, Ioannis Patras, James Oldfield, Jiankang Deng, Markos Georgopoulos, Mihalis A. Nicolaou, Yannis Panagakis","submitted_at":"2024-02-19T21:20:22Z","abstract_excerpt":"The Mixture of Experts (MoE) paradigm provides a powerful way to decompose dense layers into smaller, modular computations often more amenable to human interpretation, debugging, and editability. However, a major challenge lies in the computational cost of scaling the number of experts high enough to achieve fine-grained specialization. In this paper, we propose the Multilinear Mixture of Experts ($\\mu$MoE) layer to address this, focusing on vision models. $\\mu$MoE layers enable scalable expert specialization by performing an implicit computation on prohibitively large weight tensors entirely "},"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":"2402.12550","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-02-19T21:20:22Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"6bde6abe6262db1a3b762829f682b43fdc5f22a09cf48c516cf91249c3eb23e3","abstract_canon_sha256":"f1aa9abbf23736aef03e0d3a1b67cab3d3916ca573e68fc71e36f613defa1028"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:21:41.688771Z","signature_b64":"cRHkeqzPKpwhSpnRvbCEiOUcqmYWoEghIiE2GNmb7ru9+lwLYDTU7ehmzGtLoeNDlndfeb/5JVyUORKCk3KODA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f6d658ec72b3831cb006d839eb34e8002c8bfa0c98da76c29372ba5645a5f818","last_reissued_at":"2026-07-05T09:21:41.688237Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:21:41.688237Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multilinear Mixture of Experts: Scalable Expert Specialization through Factorization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Christos Tzelepis, Grigorios G. Chrysos, Ioannis Patras, James Oldfield, Jiankang Deng, Markos Georgopoulos, Mihalis A. Nicolaou, Yannis Panagakis","submitted_at":"2024-02-19T21:20:22Z","abstract_excerpt":"The Mixture of Experts (MoE) paradigm provides a powerful way to decompose dense layers into smaller, modular computations often more amenable to human interpretation, debugging, and editability. However, a major challenge lies in the computational cost of scaling the number of experts high enough to achieve fine-grained specialization. In this paper, we propose the Multilinear Mixture of Experts ($\\mu$MoE) layer to address this, focusing on vision models. $\\mu$MoE layers enable scalable expert specialization by performing an implicit computation on prohibitively large weight tensors entirely "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.12550","kind":"arxiv","version":4},"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/2402.12550/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":"2402.12550","created_at":"2026-07-05T09:21:41.688299+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.12550v4","created_at":"2026-07-05T09:21:41.688299+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.12550","created_at":"2026-07-05T09:21:41.688299+00:00"},{"alias_kind":"pith_short_12","alias_value":"63LFR3DSWOBR","created_at":"2026-07-05T09:21:41.688299+00:00"},{"alias_kind":"pith_short_16","alias_value":"63LFR3DSWOBRZMAG","created_at":"2026-07-05T09:21:41.688299+00:00"},{"alias_kind":"pith_short_8","alias_value":"63LFR3DS","created_at":"2026-07-05T09:21:41.688299+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2509.08032","citing_title":"SciGPT: A Large Language Model for Scientific Literature Understanding and Knowledge Discovery","ref_index":12,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/63LFR3DSWOBRZMAG3A46WNHIAA","json":"https://pith.science/pith/63LFR3DSWOBRZMAG3A46WNHIAA.json","graph_json":"https://pith.science/api/pith-number/63LFR3DSWOBRZMAG3A46WNHIAA/graph.json","events_json":"https://pith.science/api/pith-number/63LFR3DSWOBRZMAG3A46WNHIAA/events.json","paper":"https://pith.science/paper/63LFR3DS"},"agent_actions":{"view_html":"https://pith.science/pith/63LFR3DSWOBRZMAG3A46WNHIAA","download_json":"https://pith.science/pith/63LFR3DSWOBRZMAG3A46WNHIAA.json","view_paper":"https://pith.science/paper/63LFR3DS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.12550&json=true","fetch_graph":"https://pith.science/api/pith-number/63LFR3DSWOBRZMAG3A46WNHIAA/graph.json","fetch_events":"https://pith.science/api/pith-number/63LFR3DSWOBRZMAG3A46WNHIAA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/63LFR3DSWOBRZMAG3A46WNHIAA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/63LFR3DSWOBRZMAG3A46WNHIAA/action/storage_attestation","attest_author":"https://pith.science/pith/63LFR3DSWOBRZMAG3A46WNHIAA/action/author_attestation","sign_citation":"https://pith.science/pith/63LFR3DSWOBRZMAG3A46WNHIAA/action/citation_signature","submit_replication":"https://pith.science/pith/63LFR3DSWOBRZMAG3A46WNHIAA/action/replication_record"}},"created_at":"2026-07-05T09:21:41.688299+00:00","updated_at":"2026-07-05T09:21:41.688299+00:00"}