{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:JTPPH6IYRVKLOWHVOWICA3TEXN","short_pith_number":"pith:JTPPH6IY","schema_version":"1.0","canonical_sha256":"4cdef3f9188d54b758f57590206e64bb60227b38efafe0520db38ddcacdda3df","source":{"kind":"arxiv","id":"2607.10420","version":1},"attestation_state":"computed","paper":{"title":"Is Model Instability just Noise to be Tolerated or a Property that can be Managed?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.SE","authors_text":"Amirali Rayegan, Lunxiao Li, Tim Menzies","submitted_at":"2026-07-11T17:56:46Z","abstract_excerpt":"In software analytics, rerunning the same analysis twice often yields different models and conclusions. This reduces trust in the model and limits its use. We find that model instability is a major problem. Across 127 multi-objective SE optimization problems (12,700 test cases), repeated runs of a state-of-the-art optimizer agree on only 13.7% of test cases, even under improved settings. We argue that this instability is not merely noise to tolerate, but a property that can be measured and managed. By adjusting how labels are spent, how complex the models become, and how splits are scored, we "},"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":"2607.10420","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-07-11T17:56:46Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"3b0b056505027790536a4d60f4f1b0a861dc1906d3e682f3ab8369dc08d60ed9","abstract_canon_sha256":"be315e82065ac78a0a51a9bfd412492980840b34706e6aca1bc8fffa83a852c9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-14T01:21:15.152090Z","signature_b64":"2GdxJk8eMqK+sq3msrT2RcFW+yWqiJPCqwRhKlBKMebTkTCKSfSyMr5MeSppoYRZOKwSSUwqIcabdXGvbKJNAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4cdef3f9188d54b758f57590206e64bb60227b38efafe0520db38ddcacdda3df","last_reissued_at":"2026-07-14T01:21:15.151266Z","signature_status":"signed_v1","first_computed_at":"2026-07-14T01:21:15.151266Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Is Model Instability just Noise to be Tolerated or a Property that can be Managed?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.SE","authors_text":"Amirali Rayegan, Lunxiao Li, Tim Menzies","submitted_at":"2026-07-11T17:56:46Z","abstract_excerpt":"In software analytics, rerunning the same analysis twice often yields different models and conclusions. This reduces trust in the model and limits its use. We find that model instability is a major problem. Across 127 multi-objective SE optimization problems (12,700 test cases), repeated runs of a state-of-the-art optimizer agree on only 13.7% of test cases, even under improved settings. We argue that this instability is not merely noise to tolerate, but a property that can be measured and managed. By adjusting how labels are spent, how complex the models become, and how splits are scored, we "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.10420","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/2607.10420/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":"2607.10420","created_at":"2026-07-14T01:21:15.151698+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.10420v1","created_at":"2026-07-14T01:21:15.151698+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.10420","created_at":"2026-07-14T01:21:15.151698+00:00"},{"alias_kind":"pith_short_12","alias_value":"JTPPH6IYRVKL","created_at":"2026-07-14T01:21:15.151698+00:00"},{"alias_kind":"pith_short_16","alias_value":"JTPPH6IYRVKLOWHV","created_at":"2026-07-14T01:21:15.151698+00:00"},{"alias_kind":"pith_short_8","alias_value":"JTPPH6IY","created_at":"2026-07-14T01:21:15.151698+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/JTPPH6IYRVKLOWHVOWICA3TEXN","json":"https://pith.science/pith/JTPPH6IYRVKLOWHVOWICA3TEXN.json","graph_json":"https://pith.science/api/pith-number/JTPPH6IYRVKLOWHVOWICA3TEXN/graph.json","events_json":"https://pith.science/api/pith-number/JTPPH6IYRVKLOWHVOWICA3TEXN/events.json","paper":"https://pith.science/paper/JTPPH6IY"},"agent_actions":{"view_html":"https://pith.science/pith/JTPPH6IYRVKLOWHVOWICA3TEXN","download_json":"https://pith.science/pith/JTPPH6IYRVKLOWHVOWICA3TEXN.json","view_paper":"https://pith.science/paper/JTPPH6IY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.10420&json=true","fetch_graph":"https://pith.science/api/pith-number/JTPPH6IYRVKLOWHVOWICA3TEXN/graph.json","fetch_events":"https://pith.science/api/pith-number/JTPPH6IYRVKLOWHVOWICA3TEXN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JTPPH6IYRVKLOWHVOWICA3TEXN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JTPPH6IYRVKLOWHVOWICA3TEXN/action/storage_attestation","attest_author":"https://pith.science/pith/JTPPH6IYRVKLOWHVOWICA3TEXN/action/author_attestation","sign_citation":"https://pith.science/pith/JTPPH6IYRVKLOWHVOWICA3TEXN/action/citation_signature","submit_replication":"https://pith.science/pith/JTPPH6IYRVKLOWHVOWICA3TEXN/action/replication_record"}},"created_at":"2026-07-14T01:21:15.151698+00:00","updated_at":"2026-07-14T01:21:15.151698+00:00"}