{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:BQQSHGSOUCPMX6P7UGVWK74WBB","short_pith_number":"pith:BQQSHGSO","schema_version":"1.0","canonical_sha256":"0c21239a4ea09ecbf9ffa1ab657f96086e3a1e7ccd2d4eadd9f567226a03d967","source":{"kind":"arxiv","id":"2406.07529","version":5},"attestation_state":"computed","paper":{"title":"MAP: Low-compute Model Merging with Amortized Pareto Fronts via Quadratic Approximation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Huan He, Jiang Bian, Jie Fu, Lu Li, Suyuchen Wang, Tianyu Zhang, Yong Chen, Yonghui Wu, Yoshua Bengio, Zhiqi Bu","submitted_at":"2024-06-11T17:55:25Z","abstract_excerpt":"Model merging has emerged as an effective approach to combine multiple single-task models into a multitask model. This process typically involves computing a weighted average of the model parameters without any additional training. Existing model-merging methods focus on enhancing average task accuracy. However, interference and conflicts between the objectives of different tasks can lead to trade-offs during the merging process. In real-world applications, a set of solutions with various trade-offs can be more informative, helping practitioners make decisions based on diverse preferences. In "},"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":"2406.07529","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-11T17:55:25Z","cross_cats_sorted":[],"title_canon_sha256":"45cf169727c406ecbbc634a550ed51681f56fbd34e505a324a831383c654cc56","abstract_canon_sha256":"af48dc1274c8fb23cc58b2838038fc9ac16bc98a8df39ab225160c20fd58c65f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:53:45.586033Z","signature_b64":"4rAuMGg5kph/1JqH3EmWhCyqZVA4FKZ4DletBrxaucV3FCQzmc8D2v4MuyKsVjhiLCXhiItBSO5SFP1GeVa6DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0c21239a4ea09ecbf9ffa1ab657f96086e3a1e7ccd2d4eadd9f567226a03d967","last_reissued_at":"2026-07-05T10:53:45.585502Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:53:45.585502Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MAP: Low-compute Model Merging with Amortized Pareto Fronts via Quadratic Approximation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Huan He, Jiang Bian, Jie Fu, Lu Li, Suyuchen Wang, Tianyu Zhang, Yong Chen, Yonghui Wu, Yoshua Bengio, Zhiqi Bu","submitted_at":"2024-06-11T17:55:25Z","abstract_excerpt":"Model merging has emerged as an effective approach to combine multiple single-task models into a multitask model. This process typically involves computing a weighted average of the model parameters without any additional training. Existing model-merging methods focus on enhancing average task accuracy. However, interference and conflicts between the objectives of different tasks can lead to trade-offs during the merging process. In real-world applications, a set of solutions with various trade-offs can be more informative, helping practitioners make decisions based on diverse preferences. In "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.07529","kind":"arxiv","version":5},"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/2406.07529/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":"2406.07529","created_at":"2026-07-05T10:53:45.585566+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.07529v5","created_at":"2026-07-05T10:53:45.585566+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.07529","created_at":"2026-07-05T10:53:45.585566+00:00"},{"alias_kind":"pith_short_12","alias_value":"BQQSHGSOUCPM","created_at":"2026-07-05T10:53:45.585566+00:00"},{"alias_kind":"pith_short_16","alias_value":"BQQSHGSOUCPMX6P7","created_at":"2026-07-05T10:53:45.585566+00:00"},{"alias_kind":"pith_short_8","alias_value":"BQQSHGSO","created_at":"2026-07-05T10:53:45.585566+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.01689","citing_title":"Model Merging as Probabilistic Inference in Fine-Tuning Parameter Space","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2408.07666","citing_title":"Model Merging in LLMs, MLLMs, and Beyond: Methods, Theories, Applications and Opportunities","ref_index":120,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14546","citing_title":"Discovering Physical Directions in Weight Space: Composing Neural PDE Experts","ref_index":59,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BQQSHGSOUCPMX6P7UGVWK74WBB","json":"https://pith.science/pith/BQQSHGSOUCPMX6P7UGVWK74WBB.json","graph_json":"https://pith.science/api/pith-number/BQQSHGSOUCPMX6P7UGVWK74WBB/graph.json","events_json":"https://pith.science/api/pith-number/BQQSHGSOUCPMX6P7UGVWK74WBB/events.json","paper":"https://pith.science/paper/BQQSHGSO"},"agent_actions":{"view_html":"https://pith.science/pith/BQQSHGSOUCPMX6P7UGVWK74WBB","download_json":"https://pith.science/pith/BQQSHGSOUCPMX6P7UGVWK74WBB.json","view_paper":"https://pith.science/paper/BQQSHGSO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.07529&json=true","fetch_graph":"https://pith.science/api/pith-number/BQQSHGSOUCPMX6P7UGVWK74WBB/graph.json","fetch_events":"https://pith.science/api/pith-number/BQQSHGSOUCPMX6P7UGVWK74WBB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BQQSHGSOUCPMX6P7UGVWK74WBB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BQQSHGSOUCPMX6P7UGVWK74WBB/action/storage_attestation","attest_author":"https://pith.science/pith/BQQSHGSOUCPMX6P7UGVWK74WBB/action/author_attestation","sign_citation":"https://pith.science/pith/BQQSHGSOUCPMX6P7UGVWK74WBB/action/citation_signature","submit_replication":"https://pith.science/pith/BQQSHGSOUCPMX6P7UGVWK74WBB/action/replication_record"}},"created_at":"2026-07-05T10:53:45.585566+00:00","updated_at":"2026-07-05T10:53:45.585566+00:00"}