{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:HLJXRV4GHX3XPAHPQL5QWID7FZ","short_pith_number":"pith:HLJXRV4G","schema_version":"1.0","canonical_sha256":"3ad378d7863df77780ef82fb0b207f2e4e3aecef2cb76ec6cfb697902d4d4f4a","source":{"kind":"arxiv","id":"2402.12656","version":4},"attestation_state":"computed","paper":{"title":"HyperMoE: Towards Better Mixture of Experts via Transferring Among Experts","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Hao Zhao, Huijia Wu, Jie Fu, Zhaofeng He, Zihan Qiu, Zili Wang","submitted_at":"2024-02-20T02:09:55Z","abstract_excerpt":"The Mixture of Experts (MoE) for language models has been proven effective in augmenting the capacity of models by dynamically routing each input token to a specific subset of experts for processing. Despite the success, most existing methods face a challenge for balance between sparsity and the availability of expert knowledge: enhancing performance through increased use of expert knowledge often results in diminishing sparsity during expert selection. To mitigate this contradiction, we propose HyperMoE, a novel MoE framework built upon Hypernetworks. This framework integrates the computation"},"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":"2402.12656","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-20T02:09:55Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"cfe18ff8b98bd3c248559044590f4ea1de943aa33cd4a4b64b8a9780ebb36f32","abstract_canon_sha256":"4b08a3d0eed548adb5618f968f63f83f4cb4373b58affe9da287035550898e76"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:48:12.525621Z","signature_b64":"BLnHyh9LgXj9kLe2amvWkWeaDv3H6iMivG+HSEeJuBSBjrsgHOUZ/TBUwTI84gN3E6W60JruyBvqvB1mJcopCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3ad378d7863df77780ef82fb0b207f2e4e3aecef2cb76ec6cfb697902d4d4f4a","last_reissued_at":"2026-07-05T08:48:12.525154Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:48:12.525154Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"HyperMoE: Towards Better Mixture of Experts via Transferring Among Experts","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Hao Zhao, Huijia Wu, Jie Fu, Zhaofeng He, Zihan Qiu, Zili Wang","submitted_at":"2024-02-20T02:09:55Z","abstract_excerpt":"The Mixture of Experts (MoE) for language models has been proven effective in augmenting the capacity of models by dynamically routing each input token to a specific subset of experts for processing. Despite the success, most existing methods face a challenge for balance between sparsity and the availability of expert knowledge: enhancing performance through increased use of expert knowledge often results in diminishing sparsity during expert selection. To mitigate this contradiction, we propose HyperMoE, a novel MoE framework built upon Hypernetworks. This framework integrates the computation"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.12656","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/2402.12656/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":"2402.12656","created_at":"2026-07-05T08:48:12.525213+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.12656v4","created_at":"2026-07-05T08:48:12.525213+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.12656","created_at":"2026-07-05T08:48:12.525213+00:00"},{"alias_kind":"pith_short_12","alias_value":"HLJXRV4GHX3X","created_at":"2026-07-05T08:48:12.525213+00:00"},{"alias_kind":"pith_short_16","alias_value":"HLJXRV4GHX3XPAHP","created_at":"2026-07-05T08:48:12.525213+00:00"},{"alias_kind":"pith_short_8","alias_value":"HLJXRV4G","created_at":"2026-07-05T08:48:12.525213+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.05538","citing_title":"Less is MoE: Trimming Experts in Domain-Specialist Language Models","ref_index":51,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08322","citing_title":"SDG-MoE: Signed Debate Graph Mixture-of-Experts","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08322","citing_title":"SDG-MoE: Signed Debate Graph Mixture-of-Experts","ref_index":12,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HLJXRV4GHX3XPAHPQL5QWID7FZ","json":"https://pith.science/pith/HLJXRV4GHX3XPAHPQL5QWID7FZ.json","graph_json":"https://pith.science/api/pith-number/HLJXRV4GHX3XPAHPQL5QWID7FZ/graph.json","events_json":"https://pith.science/api/pith-number/HLJXRV4GHX3XPAHPQL5QWID7FZ/events.json","paper":"https://pith.science/paper/HLJXRV4G"},"agent_actions":{"view_html":"https://pith.science/pith/HLJXRV4GHX3XPAHPQL5QWID7FZ","download_json":"https://pith.science/pith/HLJXRV4GHX3XPAHPQL5QWID7FZ.json","view_paper":"https://pith.science/paper/HLJXRV4G","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.12656&json=true","fetch_graph":"https://pith.science/api/pith-number/HLJXRV4GHX3XPAHPQL5QWID7FZ/graph.json","fetch_events":"https://pith.science/api/pith-number/HLJXRV4GHX3XPAHPQL5QWID7FZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HLJXRV4GHX3XPAHPQL5QWID7FZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HLJXRV4GHX3XPAHPQL5QWID7FZ/action/storage_attestation","attest_author":"https://pith.science/pith/HLJXRV4GHX3XPAHPQL5QWID7FZ/action/author_attestation","sign_citation":"https://pith.science/pith/HLJXRV4GHX3XPAHPQL5QWID7FZ/action/citation_signature","submit_replication":"https://pith.science/pith/HLJXRV4GHX3XPAHPQL5QWID7FZ/action/replication_record"}},"created_at":"2026-07-05T08:48:12.525213+00:00","updated_at":"2026-07-05T08:48:12.525213+00:00"}