{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:FZYEO5AM3C72BR6NU6ZMXEONK7","short_pith_number":"pith:FZYEO5AM","schema_version":"1.0","canonical_sha256":"2e7047740cd8bfa0c7cda7b2cb91cd57c1c73ea6457ae7180d24ff122af55b12","source":{"kind":"arxiv","id":"2201.10890","version":4},"attestation_state":"computed","paper":{"title":"One Student Knows All Experts Know: From Sparse to Dense","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.CV"],"primary_cat":"cs.LG","authors_text":"Fuzhao Xue, Xiaoxin He, Xiaozhe Ren, Yang You, Yuxuan Lou","submitted_at":"2022-01-26T12:11:02Z","abstract_excerpt":"Human education system trains one student by multiple experts. Mixture-of-experts (MoE) is a powerful sparse architecture including multiple experts. However, sparse MoE model is easy to overfit, hard to deploy, and not hardware-friendly for practitioners. In this work, inspired by the human education model, we propose a novel task, knowledge integration, to obtain a dense student model (OneS) as knowledgeable as one sparse MoE. We investigate this task by proposing a general training framework including knowledge gathering and knowledge distillation. Specifically, to gather key knowledge from"},"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":"2201.10890","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-01-26T12:11:02Z","cross_cats_sorted":["cs.AI","cs.CL","cs.CV"],"title_canon_sha256":"1cf4047ab2c8fcb49bfaaf6a40c9502c0104dfa2d7f5d06e76fb1d76f7568b32","abstract_canon_sha256":"3d169844849a1ae9ebeac1d7b22e436363b554862fdb97e607f87698791f3c2b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:09:52.522092Z","signature_b64":"tjgxRWeiDKxb4gib6aWzmZRMu2X48WqxJZ3UYZX7tBSsR3GxoQaWqHihDQQpOq9qvpp8auT9nNk1jYildmVyAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2e7047740cd8bfa0c7cda7b2cb91cd57c1c73ea6457ae7180d24ff122af55b12","last_reissued_at":"2026-07-05T05:09:52.521742Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:09:52.521742Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"One Student Knows All Experts Know: From Sparse to Dense","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.CV"],"primary_cat":"cs.LG","authors_text":"Fuzhao Xue, Xiaoxin He, Xiaozhe Ren, Yang You, Yuxuan Lou","submitted_at":"2022-01-26T12:11:02Z","abstract_excerpt":"Human education system trains one student by multiple experts. Mixture-of-experts (MoE) is a powerful sparse architecture including multiple experts. However, sparse MoE model is easy to overfit, hard to deploy, and not hardware-friendly for practitioners. In this work, inspired by the human education model, we propose a novel task, knowledge integration, to obtain a dense student model (OneS) as knowledgeable as one sparse MoE. We investigate this task by proposing a general training framework including knowledge gathering and knowledge distillation. Specifically, to gather key knowledge from"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2201.10890","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/2201.10890/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":"2201.10890","created_at":"2026-07-05T05:09:52.521797+00:00"},{"alias_kind":"arxiv_version","alias_value":"2201.10890v4","created_at":"2026-07-05T05:09:52.521797+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2201.10890","created_at":"2026-07-05T05:09:52.521797+00:00"},{"alias_kind":"pith_short_12","alias_value":"FZYEO5AM3C72","created_at":"2026-07-05T05:09:52.521797+00:00"},{"alias_kind":"pith_short_16","alias_value":"FZYEO5AM3C72BR6N","created_at":"2026-07-05T05:09:52.521797+00:00"},{"alias_kind":"pith_short_8","alias_value":"FZYEO5AM","created_at":"2026-07-05T05:09:52.521797+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.01709","citing_title":"MoVE-KD: Knowledge Distillation for VLMs with Mixture of Visual Encoders","ref_index":46,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FZYEO5AM3C72BR6NU6ZMXEONK7","json":"https://pith.science/pith/FZYEO5AM3C72BR6NU6ZMXEONK7.json","graph_json":"https://pith.science/api/pith-number/FZYEO5AM3C72BR6NU6ZMXEONK7/graph.json","events_json":"https://pith.science/api/pith-number/FZYEO5AM3C72BR6NU6ZMXEONK7/events.json","paper":"https://pith.science/paper/FZYEO5AM"},"agent_actions":{"view_html":"https://pith.science/pith/FZYEO5AM3C72BR6NU6ZMXEONK7","download_json":"https://pith.science/pith/FZYEO5AM3C72BR6NU6ZMXEONK7.json","view_paper":"https://pith.science/paper/FZYEO5AM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2201.10890&json=true","fetch_graph":"https://pith.science/api/pith-number/FZYEO5AM3C72BR6NU6ZMXEONK7/graph.json","fetch_events":"https://pith.science/api/pith-number/FZYEO5AM3C72BR6NU6ZMXEONK7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FZYEO5AM3C72BR6NU6ZMXEONK7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FZYEO5AM3C72BR6NU6ZMXEONK7/action/storage_attestation","attest_author":"https://pith.science/pith/FZYEO5AM3C72BR6NU6ZMXEONK7/action/author_attestation","sign_citation":"https://pith.science/pith/FZYEO5AM3C72BR6NU6ZMXEONK7/action/citation_signature","submit_replication":"https://pith.science/pith/FZYEO5AM3C72BR6NU6ZMXEONK7/action/replication_record"}},"created_at":"2026-07-05T05:09:52.521797+00:00","updated_at":"2026-07-05T05:09:52.521797+00:00"}