{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:5VTGOBVZQKZTTGAZEKF3FYA4MH","short_pith_number":"pith:5VTGOBVZ","schema_version":"1.0","canonical_sha256":"ed666706b982b3399819228bb2e01c61fa130e9cfb16cdcb758a41a69b4a68d6","source":{"kind":"arxiv","id":"2412.17023","version":1},"attestation_state":"computed","paper":{"title":"Parameter-Efficient Interventions for Enhanced Model Merging","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bartosz Zieli\\'nski, Daniel Marczak, Marcin Osial","submitted_at":"2024-12-22T13:58:12Z","abstract_excerpt":"Model merging combines knowledge from task-specific models into a unified multi-task model to avoid joint training on all task data. However, current methods face challenges due to representation bias, which can interfere with tasks performance. As a remedy, we propose IntervMerge, a novel approach to multi-task model merging that effectively mitigates representation bias across the model using taskspecific interventions. To further enhance its efficiency, we introduce mini-interventions, which modify only part of the representation, thereby reducing the additional parameters without compromis"},"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":"2412.17023","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-12-22T13:58:12Z","cross_cats_sorted":[],"title_canon_sha256":"8262e4964034b4dece53bbba5fa8835c775a54b66558921c3decd83cdf361736","abstract_canon_sha256":"1957ddafb0f931d1e1081c481e7174bd3f42d8227118fd6f03c5c28d0e151159"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:53:16.744207Z","signature_b64":"NfiNonjcI40+/AKZaYH8Tk/8++JpjYZ0fJVX63Hv1+X6LP3F1JFMAkhM82fwboGyRHj7NIDlh061GSmnlZ4QAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ed666706b982b3399819228bb2e01c61fa130e9cfb16cdcb758a41a69b4a68d6","last_reissued_at":"2026-07-05T09:53:16.743836Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:53:16.743836Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Parameter-Efficient Interventions for Enhanced Model Merging","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bartosz Zieli\\'nski, Daniel Marczak, Marcin Osial","submitted_at":"2024-12-22T13:58:12Z","abstract_excerpt":"Model merging combines knowledge from task-specific models into a unified multi-task model to avoid joint training on all task data. However, current methods face challenges due to representation bias, which can interfere with tasks performance. As a remedy, we propose IntervMerge, a novel approach to multi-task model merging that effectively mitigates representation bias across the model using taskspecific interventions. To further enhance its efficiency, we introduce mini-interventions, which modify only part of the representation, thereby reducing the additional parameters without compromis"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.17023","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/2412.17023/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":"2412.17023","created_at":"2026-07-05T09:53:16.743885+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.17023v1","created_at":"2026-07-05T09:53:16.743885+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.17023","created_at":"2026-07-05T09:53:16.743885+00:00"},{"alias_kind":"pith_short_12","alias_value":"5VTGOBVZQKZT","created_at":"2026-07-05T09:53:16.743885+00:00"},{"alias_kind":"pith_short_16","alias_value":"5VTGOBVZQKZTTGAZ","created_at":"2026-07-05T09:53:16.743885+00:00"},{"alias_kind":"pith_short_8","alias_value":"5VTGOBVZ","created_at":"2026-07-05T09:53:16.743885+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/5VTGOBVZQKZTTGAZEKF3FYA4MH","json":"https://pith.science/pith/5VTGOBVZQKZTTGAZEKF3FYA4MH.json","graph_json":"https://pith.science/api/pith-number/5VTGOBVZQKZTTGAZEKF3FYA4MH/graph.json","events_json":"https://pith.science/api/pith-number/5VTGOBVZQKZTTGAZEKF3FYA4MH/events.json","paper":"https://pith.science/paper/5VTGOBVZ"},"agent_actions":{"view_html":"https://pith.science/pith/5VTGOBVZQKZTTGAZEKF3FYA4MH","download_json":"https://pith.science/pith/5VTGOBVZQKZTTGAZEKF3FYA4MH.json","view_paper":"https://pith.science/paper/5VTGOBVZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.17023&json=true","fetch_graph":"https://pith.science/api/pith-number/5VTGOBVZQKZTTGAZEKF3FYA4MH/graph.json","fetch_events":"https://pith.science/api/pith-number/5VTGOBVZQKZTTGAZEKF3FYA4MH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5VTGOBVZQKZTTGAZEKF3FYA4MH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5VTGOBVZQKZTTGAZEKF3FYA4MH/action/storage_attestation","attest_author":"https://pith.science/pith/5VTGOBVZQKZTTGAZEKF3FYA4MH/action/author_attestation","sign_citation":"https://pith.science/pith/5VTGOBVZQKZTTGAZEKF3FYA4MH/action/citation_signature","submit_replication":"https://pith.science/pith/5VTGOBVZQKZTTGAZEKF3FYA4MH/action/replication_record"}},"created_at":"2026-07-05T09:53:16.743885+00:00","updated_at":"2026-07-05T09:53:16.743885+00:00"}