{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:47K7IP3NAX23PLCVUHRIKUR3T3","short_pith_number":"pith:47K7IP3N","schema_version":"1.0","canonical_sha256":"e7d5f43f6d05f5b7ac55a1e285523b9ec6fbbc2a76ed6d8eca52067ba3ed72f3","source":{"kind":"arxiv","id":"2407.06322","version":2},"attestation_state":"computed","paper":{"title":"MagMax: Leveraging Model Merging for Seamless Continual Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Bart{\\l}omiej Twardowski, Daniel Marczak, Sebastian Cygert, Tomasz Trzci\\'nski","submitted_at":"2024-07-08T18:38:52Z","abstract_excerpt":"This paper introduces a continual learning approach named MagMax, which utilizes model merging to enable large pre-trained models to continuously learn from new data without forgetting previously acquired knowledge. Distinct from traditional continual learning methods that aim to reduce forgetting during task training, MagMax combines sequential fine-tuning with a maximum magnitude weight selection for effective knowledge integration across tasks. Our initial contribution is an extensive examination of model merging techniques, revealing that simple approaches like weight averaging and random "},"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":"2407.06322","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-07-08T18:38:52Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"e016242245d298adb57383e1c2cb04a9844416d34798bfce4386c46b7f4d02dc","abstract_canon_sha256":"d0d4e421a247c32b06b1ecc11571ce1e5b2c7ec8130ca935adbf7011ad58c94b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:49:53.976084Z","signature_b64":"kb00t5twl4ObY53iIhF8x2EMqKUW/SB0oegFdgg0/3Rho29j8ptDXQ1DmC/5rpg10NBxr5Sqy4oT0zHF8IFQAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e7d5f43f6d05f5b7ac55a1e285523b9ec6fbbc2a76ed6d8eca52067ba3ed72f3","last_reissued_at":"2026-07-05T08:49:53.975738Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:49:53.975738Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MagMax: Leveraging Model Merging for Seamless Continual Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Bart{\\l}omiej Twardowski, Daniel Marczak, Sebastian Cygert, Tomasz Trzci\\'nski","submitted_at":"2024-07-08T18:38:52Z","abstract_excerpt":"This paper introduces a continual learning approach named MagMax, which utilizes model merging to enable large pre-trained models to continuously learn from new data without forgetting previously acquired knowledge. Distinct from traditional continual learning methods that aim to reduce forgetting during task training, MagMax combines sequential fine-tuning with a maximum magnitude weight selection for effective knowledge integration across tasks. Our initial contribution is an extensive examination of model merging techniques, revealing that simple approaches like weight averaging and random "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.06322","kind":"arxiv","version":2},"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/2407.06322/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":"2407.06322","created_at":"2026-07-05T08:49:53.975791+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.06322v2","created_at":"2026-07-05T08:49:53.975791+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.06322","created_at":"2026-07-05T08:49:53.975791+00:00"},{"alias_kind":"pith_short_12","alias_value":"47K7IP3NAX23","created_at":"2026-07-05T08:49:53.975791+00:00"},{"alias_kind":"pith_short_16","alias_value":"47K7IP3NAX23PLCV","created_at":"2026-07-05T08:49:53.975791+00:00"},{"alias_kind":"pith_short_8","alias_value":"47K7IP3N","created_at":"2026-07-05T08:49:53.975791+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/47K7IP3NAX23PLCVUHRIKUR3T3","json":"https://pith.science/pith/47K7IP3NAX23PLCVUHRIKUR3T3.json","graph_json":"https://pith.science/api/pith-number/47K7IP3NAX23PLCVUHRIKUR3T3/graph.json","events_json":"https://pith.science/api/pith-number/47K7IP3NAX23PLCVUHRIKUR3T3/events.json","paper":"https://pith.science/paper/47K7IP3N"},"agent_actions":{"view_html":"https://pith.science/pith/47K7IP3NAX23PLCVUHRIKUR3T3","download_json":"https://pith.science/pith/47K7IP3NAX23PLCVUHRIKUR3T3.json","view_paper":"https://pith.science/paper/47K7IP3N","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.06322&json=true","fetch_graph":"https://pith.science/api/pith-number/47K7IP3NAX23PLCVUHRIKUR3T3/graph.json","fetch_events":"https://pith.science/api/pith-number/47K7IP3NAX23PLCVUHRIKUR3T3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/47K7IP3NAX23PLCVUHRIKUR3T3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/47K7IP3NAX23PLCVUHRIKUR3T3/action/storage_attestation","attest_author":"https://pith.science/pith/47K7IP3NAX23PLCVUHRIKUR3T3/action/author_attestation","sign_citation":"https://pith.science/pith/47K7IP3NAX23PLCVUHRIKUR3T3/action/citation_signature","submit_replication":"https://pith.science/pith/47K7IP3NAX23PLCVUHRIKUR3T3/action/replication_record"}},"created_at":"2026-07-05T08:49:53.975791+00:00","updated_at":"2026-07-05T08:49:53.975791+00:00"}