{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:G5PHVFIDGAADWC23K6NVBP63LN","short_pith_number":"pith:G5PHVFID","schema_version":"1.0","canonical_sha256":"375e7a950330003b0b5b579b50bfdb5b67b3a269208585bf5dd3c0aa84c7e4d7","source":{"kind":"arxiv","id":"2601.22285","version":7},"attestation_state":"computed","paper":{"title":"Demystifying Mergeability: Interpretable Properties to Predict Model Merging Success","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Gradient alignment between fine-tuned models is the strongest predictor of successful merging across methods and tasks.","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Bo Zhao, Emanuele Rodol\\`a, Luca Zhou, Rose Yu","submitted_at":"2026-01-29T20:00:26Z","abstract_excerpt":"Model merging combines knowledge from separately fine-tuned models, yet the factors driving its success remain poorly understood. While recent work treats mergeability as an intrinsic property of the models, we show with an architecture-agnostic framework that it fundamentally depends on both the merging method and the partner tasks. Using L1-regularized linear optimization over a set of interpretable pairwise metrics (e.g., gradient L_2 distance), we uncover properties correlating with post-merge normalized accuracy across five merging methods. We find architecture- and method-specific variat"},"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":"2601.22285","kind":"arxiv","version":7},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-01-29T20:00:26Z","cross_cats_sorted":[],"title_canon_sha256":"f7b423489f9f9ad0a232ce96a629dcadcaf3bb5155bc709c0dc487ee8cd977f3","abstract_canon_sha256":"12d878280646b316ab674d039cd7d2ac0928643b7de47e413c5b730ca4cfa0ba"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T01:05:07.282667Z","signature_b64":"JIywg1QJMsZH8Vwh082XMDgm0ALJgxyNgA9PeRfKuerIQaW4uKwgqAWR/Q6rq8xAnuDk+ROM0vMY5tE3GAn2Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"375e7a950330003b0b5b579b50bfdb5b67b3a269208585bf5dd3c0aa84c7e4d7","last_reissued_at":"2026-05-20T01:05:07.281625Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T01:05:07.281625Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Demystifying Mergeability: Interpretable Properties to Predict Model Merging Success","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Gradient alignment between fine-tuned models is the strongest predictor of successful merging across methods and tasks.","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Bo Zhao, Emanuele Rodol\\`a, Luca Zhou, Rose Yu","submitted_at":"2026-01-29T20:00:26Z","abstract_excerpt":"Model merging combines knowledge from separately fine-tuned models, yet the factors driving its success remain poorly understood. While recent work treats mergeability as an intrinsic property of the models, we show with an architecture-agnostic framework that it fundamentally depends on both the merging method and the partner tasks. Using L1-regularized linear optimization over a set of interpretable pairwise metrics (e.g., gradient L_2 distance), we uncover properties correlating with post-merge normalized accuracy across five merging methods. We find architecture- and method-specific variat"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Crucially, however, gradient alignment metrics consistently emerge as the most fundamental signals of compatibility.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the chosen set of interpretable pairwise metrics (such as gradient L2 distance) plus L1-regularized linear optimization is sufficient to identify the true drivers of merge success without missing key unmeasured factors.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Mergeability is not intrinsic to models but depends on merging method and tasks, with gradient alignment as the key predictive signal uncovered via interpretable metrics.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Gradient alignment between fine-tuned models is the strongest predictor of successful merging across methods and tasks.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"00cb05c533c0e179389859adb011782384da3ab79ca3fe96afc3fd518f657d7d"},"source":{"id":"2601.22285","kind":"arxiv","version":7},"verdict":{"id":"dbf9c059-828e-416f-a712-6c617498264b","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-16T09:47:05.919474Z","strongest_claim":"Crucially, however, gradient alignment metrics consistently emerge as the most fundamental signals of compatibility.","one_line_summary":"Mergeability is not intrinsic to models but depends on merging method and tasks, with gradient alignment as the key predictive signal uncovered via interpretable metrics.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the chosen set of interpretable pairwise metrics (such as gradient L2 distance) plus L1-regularized linear optimization is sufficient to identify the true drivers of merge success without missing key unmeasured factors.","pith_extraction_headline":"Gradient alignment between fine-tuned models is the strongest predictor of successful merging across methods and tasks."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2601.22285/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":"2601.22285","created_at":"2026-05-20T01:05:07.281776+00:00"},{"alias_kind":"arxiv_version","alias_value":"2601.22285v7","created_at":"2026-05-20T01:05:07.281776+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.22285","created_at":"2026-05-20T01:05:07.281776+00:00"},{"alias_kind":"pith_short_12","alias_value":"G5PHVFIDGAAD","created_at":"2026-05-20T01:05:07.281776+00:00"},{"alias_kind":"pith_short_16","alias_value":"G5PHVFIDGAADWC23","created_at":"2026-05-20T01:05:07.281776+00:00"},{"alias_kind":"pith_short_8","alias_value":"G5PHVFID","created_at":"2026-05-20T01:05:07.281776+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":3,"sample":[{"citing_arxiv_id":"2605.13652","citing_title":"Beyond Perplexity: A Geometric and Spectral Study of Low-Rank Pre-Training","ref_index":30,"is_internal_anchor":true},{"citing_arxiv_id":"2605.13652","citing_title":"Beyond Perplexity: A Geometric and Spectral Study of Low-Rank Pre-Training","ref_index":30,"is_internal_anchor":true},{"citing_arxiv_id":"2605.01580","citing_title":"Model Merging: Foundations and Algorithms","ref_index":207,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/G5PHVFIDGAADWC23K6NVBP63LN","json":"https://pith.science/pith/G5PHVFIDGAADWC23K6NVBP63LN.json","graph_json":"https://pith.science/api/pith-number/G5PHVFIDGAADWC23K6NVBP63LN/graph.json","events_json":"https://pith.science/api/pith-number/G5PHVFIDGAADWC23K6NVBP63LN/events.json","paper":"https://pith.science/paper/G5PHVFID"},"agent_actions":{"view_html":"https://pith.science/pith/G5PHVFIDGAADWC23K6NVBP63LN","download_json":"https://pith.science/pith/G5PHVFIDGAADWC23K6NVBP63LN.json","view_paper":"https://pith.science/paper/G5PHVFID","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2601.22285&json=true","fetch_graph":"https://pith.science/api/pith-number/G5PHVFIDGAADWC23K6NVBP63LN/graph.json","fetch_events":"https://pith.science/api/pith-number/G5PHVFIDGAADWC23K6NVBP63LN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/G5PHVFIDGAADWC23K6NVBP63LN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/G5PHVFIDGAADWC23K6NVBP63LN/action/storage_attestation","attest_author":"https://pith.science/pith/G5PHVFIDGAADWC23K6NVBP63LN/action/author_attestation","sign_citation":"https://pith.science/pith/G5PHVFIDGAADWC23K6NVBP63LN/action/citation_signature","submit_replication":"https://pith.science/pith/G5PHVFIDGAADWC23K6NVBP63LN/action/replication_record"}},"created_at":"2026-05-20T01:05:07.281776+00:00","updated_at":"2026-05-20T01:05:07.281776+00:00"}