{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:VWK2BUIHFUVQBNCJQNXP6XETLE","short_pith_number":"pith:VWK2BUIH","schema_version":"1.0","canonical_sha256":"ad95a0d1072d2b00b449836eff5c935903afc7619cafe7846c1cc6b937b9883d","source":{"kind":"arxiv","id":"2503.08998","version":1},"attestation_state":"computed","paper":{"title":"From Task-Specific Models to Unified Systems: A Review of Model Merging Approaches","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Jin Lu, Tianming Liu, Tianze Yang, Wei Ruan, Yifan Zhou","submitted_at":"2025-03-12T02:17:31Z","abstract_excerpt":"Model merging has achieved significant success, with numerous innovative methods proposed to enhance capabilities by combining multiple models. However, challenges persist due to the lack of a unified framework for classification and systematic comparative analysis, leading to inconsistencies in terminologies and categorizations. Meanwhile, as an increasing number of fine-tuned models are publicly available, their original training data often remain inaccessible due to privacy concerns or intellectual property restrictions. This makes traditional multi-task learning based on shared training da"},"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":"2503.08998","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-03-12T02:17:31Z","cross_cats_sorted":[],"title_canon_sha256":"94e733cb4576d036f53ec2b91c9f5bef7562262da1558ee25664c9b698769462","abstract_canon_sha256":"6fe1ade92b4724fc5a61639a96dd8bbea4dc8c0f08df258c4633ae9a77cca682"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:29:45.620366Z","signature_b64":"9fMPMLDuuAXJJjpsGnZrEgZeQw70DTEOn0A9FQeDR2dcmixdQpiZnnA7CYAYKjpviTSk4JeKaVP1U6e0rXH4Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ad95a0d1072d2b00b449836eff5c935903afc7619cafe7846c1cc6b937b9883d","last_reissued_at":"2026-07-05T10:29:45.619848Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:29:45.619848Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"From Task-Specific Models to Unified Systems: A Review of Model Merging Approaches","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Jin Lu, Tianming Liu, Tianze Yang, Wei Ruan, Yifan Zhou","submitted_at":"2025-03-12T02:17:31Z","abstract_excerpt":"Model merging has achieved significant success, with numerous innovative methods proposed to enhance capabilities by combining multiple models. However, challenges persist due to the lack of a unified framework for classification and systematic comparative analysis, leading to inconsistencies in terminologies and categorizations. Meanwhile, as an increasing number of fine-tuned models are publicly available, their original training data often remain inaccessible due to privacy concerns or intellectual property restrictions. This makes traditional multi-task learning based on shared training da"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.08998","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/2503.08998/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":"2503.08998","created_at":"2026-07-05T10:29:45.619913+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.08998v1","created_at":"2026-07-05T10:29:45.619913+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.08998","created_at":"2026-07-05T10:29:45.619913+00:00"},{"alias_kind":"pith_short_12","alias_value":"VWK2BUIHFUVQ","created_at":"2026-07-05T10:29:45.619913+00:00"},{"alias_kind":"pith_short_16","alias_value":"VWK2BUIHFUVQBNCJ","created_at":"2026-07-05T10:29:45.619913+00:00"},{"alias_kind":"pith_short_8","alias_value":"VWK2BUIH","created_at":"2026-07-05T10:29:45.619913+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.18627","citing_title":"PACT: Preserving Anchored Cores in Task-vectors for Model Merging","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2502.18036","citing_title":"Harnessing Multiple Large Language Models: A Survey on LLM Ensemble","ref_index":43,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VWK2BUIHFUVQBNCJQNXP6XETLE","json":"https://pith.science/pith/VWK2BUIHFUVQBNCJQNXP6XETLE.json","graph_json":"https://pith.science/api/pith-number/VWK2BUIHFUVQBNCJQNXP6XETLE/graph.json","events_json":"https://pith.science/api/pith-number/VWK2BUIHFUVQBNCJQNXP6XETLE/events.json","paper":"https://pith.science/paper/VWK2BUIH"},"agent_actions":{"view_html":"https://pith.science/pith/VWK2BUIHFUVQBNCJQNXP6XETLE","download_json":"https://pith.science/pith/VWK2BUIHFUVQBNCJQNXP6XETLE.json","view_paper":"https://pith.science/paper/VWK2BUIH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.08998&json=true","fetch_graph":"https://pith.science/api/pith-number/VWK2BUIHFUVQBNCJQNXP6XETLE/graph.json","fetch_events":"https://pith.science/api/pith-number/VWK2BUIHFUVQBNCJQNXP6XETLE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VWK2BUIHFUVQBNCJQNXP6XETLE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VWK2BUIHFUVQBNCJQNXP6XETLE/action/storage_attestation","attest_author":"https://pith.science/pith/VWK2BUIHFUVQBNCJQNXP6XETLE/action/author_attestation","sign_citation":"https://pith.science/pith/VWK2BUIHFUVQBNCJQNXP6XETLE/action/citation_signature","submit_replication":"https://pith.science/pith/VWK2BUIHFUVQBNCJQNXP6XETLE/action/replication_record"}},"created_at":"2026-07-05T10:29:45.619913+00:00","updated_at":"2026-07-05T10:29:45.619913+00:00"}