{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:PS4TEOBC4CKFPJZ2KU5K5VDGI2","short_pith_number":"pith:PS4TEOBC","schema_version":"1.0","canonical_sha256":"7cb9323822e09457a73a553aaed466469dd4ebdd3b7fa455d7236465fc006781","source":{"kind":"arxiv","id":"2410.21804","version":1},"attestation_state":"computed","paper":{"title":"Efficient and Effective Weight-Ensembling Mixture of Experts for Multi-Task Model Merging","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Anke Tang, Bo Du, Dacheng Tao, Enneng Yang, Guibing Guo, Lefei Zhang, Li Shen, Xiaochun Cao, Yong Luo","submitted_at":"2024-10-29T07:16:31Z","abstract_excerpt":"Multi-task learning (MTL) leverages a shared model to accomplish multiple tasks and facilitate knowledge transfer. Recent research on task arithmetic-based MTL demonstrates that merging the parameters of independently fine-tuned models can effectively achieve MTL. However, existing merging methods primarily seek a static optimal solution within the original model parameter space, which often results in performance degradation due to the inherent diversity among tasks and potential interferences. To address this challenge, in this paper, we propose a Weight-Ensembling Mixture of Experts (WEMoE)"},"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":"2410.21804","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-10-29T07:16:31Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"d3d6710af5b10372364056e6cac87646123657a3b37f148c2e1ec42f68504b0f","abstract_canon_sha256":"833d59342ce363174663eb386898a05500e8200635fe2b96c197b2e24fe1c2e9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:27:33.641094Z","signature_b64":"HU0puKcvQIwP7jhOkJZHSt0x8GWyMu6CLvzPmfQOlNaIiQZ+JBcf5pzhhuXwFuIReJVVKJZAdh9yvMoQ+yDrAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7cb9323822e09457a73a553aaed466469dd4ebdd3b7fa455d7236465fc006781","last_reissued_at":"2026-07-05T09:27:33.640630Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:27:33.640630Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Efficient and Effective Weight-Ensembling Mixture of Experts for Multi-Task Model Merging","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Anke Tang, Bo Du, Dacheng Tao, Enneng Yang, Guibing Guo, Lefei Zhang, Li Shen, Xiaochun Cao, Yong Luo","submitted_at":"2024-10-29T07:16:31Z","abstract_excerpt":"Multi-task learning (MTL) leverages a shared model to accomplish multiple tasks and facilitate knowledge transfer. Recent research on task arithmetic-based MTL demonstrates that merging the parameters of independently fine-tuned models can effectively achieve MTL. However, existing merging methods primarily seek a static optimal solution within the original model parameter space, which often results in performance degradation due to the inherent diversity among tasks and potential interferences. To address this challenge, in this paper, we propose a Weight-Ensembling Mixture of Experts (WEMoE)"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.21804","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/2410.21804/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":"2410.21804","created_at":"2026-07-05T09:27:33.640689+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.21804v1","created_at":"2026-07-05T09:27:33.640689+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.21804","created_at":"2026-07-05T09:27:33.640689+00:00"},{"alias_kind":"pith_short_12","alias_value":"PS4TEOBC4CKF","created_at":"2026-07-05T09:27:33.640689+00:00"},{"alias_kind":"pith_short_16","alias_value":"PS4TEOBC4CKFPJZ2","created_at":"2026-07-05T09:27:33.640689+00:00"},{"alias_kind":"pith_short_8","alias_value":"PS4TEOBC","created_at":"2026-07-05T09:27:33.640689+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2509.01548","citing_title":"Model Unmerging: Making Your Models Unmergeable for Secure Model Sharing","ref_index":42,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PS4TEOBC4CKFPJZ2KU5K5VDGI2","json":"https://pith.science/pith/PS4TEOBC4CKFPJZ2KU5K5VDGI2.json","graph_json":"https://pith.science/api/pith-number/PS4TEOBC4CKFPJZ2KU5K5VDGI2/graph.json","events_json":"https://pith.science/api/pith-number/PS4TEOBC4CKFPJZ2KU5K5VDGI2/events.json","paper":"https://pith.science/paper/PS4TEOBC"},"agent_actions":{"view_html":"https://pith.science/pith/PS4TEOBC4CKFPJZ2KU5K5VDGI2","download_json":"https://pith.science/pith/PS4TEOBC4CKFPJZ2KU5K5VDGI2.json","view_paper":"https://pith.science/paper/PS4TEOBC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.21804&json=true","fetch_graph":"https://pith.science/api/pith-number/PS4TEOBC4CKFPJZ2KU5K5VDGI2/graph.json","fetch_events":"https://pith.science/api/pith-number/PS4TEOBC4CKFPJZ2KU5K5VDGI2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PS4TEOBC4CKFPJZ2KU5K5VDGI2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PS4TEOBC4CKFPJZ2KU5K5VDGI2/action/storage_attestation","attest_author":"https://pith.science/pith/PS4TEOBC4CKFPJZ2KU5K5VDGI2/action/author_attestation","sign_citation":"https://pith.science/pith/PS4TEOBC4CKFPJZ2KU5K5VDGI2/action/citation_signature","submit_replication":"https://pith.science/pith/PS4TEOBC4CKFPJZ2KU5K5VDGI2/action/replication_record"}},"created_at":"2026-07-05T09:27:33.640689+00:00","updated_at":"2026-07-05T09:27:33.640689+00:00"}