{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:X3VLDDG4OLZKVLVZQBEZPFU5HX","short_pith_number":"pith:X3VLDDG4","schema_version":"1.0","canonical_sha256":"beeab18cdc72f2aaaeb9804997969d3df15e25ced4ae22a217447b2543d62367","source":{"kind":"arxiv","id":"2306.05641","version":2},"attestation_state":"computed","paper":{"title":"Toward Data Efficient Model Merging between Different Datasets without Performance Degradation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Daiki Chijiwa, Masanori Yamada, Shin'ya Yamaguchi, Tomoya Yamashita","submitted_at":"2023-06-09T03:00:34Z","abstract_excerpt":"Model merging is attracting attention as a novel method for creating a new model by combining the weights of different trained models. While previous studies reported that model merging works well for models trained on a single dataset with different random seeds, model merging between different datasets remains unsolved. In this paper, we attempt to reveal the difficulty in merging such models trained on different datasets and alleviate it. Our empirical analyses show that, in contrast to the single-dataset scenarios, dataset information needs to be accessed to achieve high accuracy when merg"},"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":"2306.05641","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-06-09T03:00:34Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"fd6593df99f95d617e65bd97db89fe37585acaf82dd2d40c824e247bb1a46274","abstract_canon_sha256":"6c8b12a600b62827ffbb4aba7bc5c3b11a7bf8215aac264f9b52bc8c767c0cdc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:09:17.010232Z","signature_b64":"PTUpERrWUyiQ/scE6Gsrgzr1b43WJEdNJ6XWJffjb6HBMG6ukzzLVoDxEYk1QytdQzPitD54xiHUzHp6nO+RAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"beeab18cdc72f2aaaeb9804997969d3df15e25ced4ae22a217447b2543d62367","last_reissued_at":"2026-07-05T09:09:17.009759Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:09:17.009759Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Toward Data Efficient Model Merging between Different Datasets without Performance Degradation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Daiki Chijiwa, Masanori Yamada, Shin'ya Yamaguchi, Tomoya Yamashita","submitted_at":"2023-06-09T03:00:34Z","abstract_excerpt":"Model merging is attracting attention as a novel method for creating a new model by combining the weights of different trained models. While previous studies reported that model merging works well for models trained on a single dataset with different random seeds, model merging between different datasets remains unsolved. In this paper, we attempt to reveal the difficulty in merging such models trained on different datasets and alleviate it. Our empirical analyses show that, in contrast to the single-dataset scenarios, dataset information needs to be accessed to achieve high accuracy when merg"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.05641","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/2306.05641/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":"2306.05641","created_at":"2026-07-05T09:09:17.009813+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.05641v2","created_at":"2026-07-05T09:09:17.009813+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.05641","created_at":"2026-07-05T09:09:17.009813+00:00"},{"alias_kind":"pith_short_12","alias_value":"X3VLDDG4OLZK","created_at":"2026-07-05T09:09:17.009813+00:00"},{"alias_kind":"pith_short_16","alias_value":"X3VLDDG4OLZKVLVZ","created_at":"2026-07-05T09:09:17.009813+00:00"},{"alias_kind":"pith_short_8","alias_value":"X3VLDDG4","created_at":"2026-07-05T09:09:17.009813+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2502.10216","citing_title":"Forget the Data and Fine-Tuning! Just Fold the Network to Compress","ref_index":82,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/X3VLDDG4OLZKVLVZQBEZPFU5HX","json":"https://pith.science/pith/X3VLDDG4OLZKVLVZQBEZPFU5HX.json","graph_json":"https://pith.science/api/pith-number/X3VLDDG4OLZKVLVZQBEZPFU5HX/graph.json","events_json":"https://pith.science/api/pith-number/X3VLDDG4OLZKVLVZQBEZPFU5HX/events.json","paper":"https://pith.science/paper/X3VLDDG4"},"agent_actions":{"view_html":"https://pith.science/pith/X3VLDDG4OLZKVLVZQBEZPFU5HX","download_json":"https://pith.science/pith/X3VLDDG4OLZKVLVZQBEZPFU5HX.json","view_paper":"https://pith.science/paper/X3VLDDG4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.05641&json=true","fetch_graph":"https://pith.science/api/pith-number/X3VLDDG4OLZKVLVZQBEZPFU5HX/graph.json","fetch_events":"https://pith.science/api/pith-number/X3VLDDG4OLZKVLVZQBEZPFU5HX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/X3VLDDG4OLZKVLVZQBEZPFU5HX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/X3VLDDG4OLZKVLVZQBEZPFU5HX/action/storage_attestation","attest_author":"https://pith.science/pith/X3VLDDG4OLZKVLVZQBEZPFU5HX/action/author_attestation","sign_citation":"https://pith.science/pith/X3VLDDG4OLZKVLVZQBEZPFU5HX/action/citation_signature","submit_replication":"https://pith.science/pith/X3VLDDG4OLZKVLVZQBEZPFU5HX/action/replication_record"}},"created_at":"2026-07-05T09:09:17.009813+00:00","updated_at":"2026-07-05T09:09:17.009813+00:00"}