{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:SKK2DPKJJ5RLPDPP2NLMYCBCXX","short_pith_number":"pith:SKK2DPKJ","schema_version":"1.0","canonical_sha256":"9295a1bd494f62b78defd356cc0822bdd0919b31698393ae7ac479223148d3f4","source":{"kind":"arxiv","id":"2607.09345","version":1},"attestation_state":"computed","paper":{"title":"How Far Are We from Detecting Flaky Tests? On the Limits of Code-Based Detection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.SE","authors_text":"Alexander Berndt, Jonathan Bell, \\\"Omer Oktay G\\\"ultekin, Sebastian Baltes, Thomas Bach","submitted_at":"2026-07-10T12:24:25Z","abstract_excerpt":"Flaky tests pass and fail on the same code version, weakening the signal of test results and disrupting continuous integration (CI) pipelines. Code-based flakiness detectors report strong benchmark results, yet their use in practice remains limited. We argue that the field is studying the wrong problem: Flakiness is not a static property of test code, which often lacks the information needed to decide whether a test is flaky. Analyzing three code-based detectors operating on test code, we found that widely used benchmarks contain shortcuts that inflate reported F1 scores and that evaluation pr"},"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":"2607.09345","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-07-10T12:24:25Z","cross_cats_sorted":[],"title_canon_sha256":"ec79e651b2babf3d9aa22c45f6d0124c19564f0e558690bf6c86ed2b716e96af","abstract_canon_sha256":"175b452318ad6361e62714c5cff923e5f8ae5bd1c0ea73172844239416462690"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-13T01:19:04.096331Z","signature_b64":"IZOIHBuuHdc0ypvDAoJl3fKqwNVlwqAqL1laU+lcHjDQLtR5BSXflft/GphVsKktbF8WbbL+z5pSu/12Mg9JBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9295a1bd494f62b78defd356cc0822bdd0919b31698393ae7ac479223148d3f4","last_reissued_at":"2026-07-13T01:19:04.095261Z","signature_status":"signed_v1","first_computed_at":"2026-07-13T01:19:04.095261Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"How Far Are We from Detecting Flaky Tests? On the Limits of Code-Based Detection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.SE","authors_text":"Alexander Berndt, Jonathan Bell, \\\"Omer Oktay G\\\"ultekin, Sebastian Baltes, Thomas Bach","submitted_at":"2026-07-10T12:24:25Z","abstract_excerpt":"Flaky tests pass and fail on the same code version, weakening the signal of test results and disrupting continuous integration (CI) pipelines. Code-based flakiness detectors report strong benchmark results, yet their use in practice remains limited. We argue that the field is studying the wrong problem: Flakiness is not a static property of test code, which often lacks the information needed to decide whether a test is flaky. Analyzing three code-based detectors operating on test code, we found that widely used benchmarks contain shortcuts that inflate reported F1 scores and that evaluation pr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.09345","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/2607.09345/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":"2607.09345","created_at":"2026-07-13T01:19:04.095815+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.09345v1","created_at":"2026-07-13T01:19:04.095815+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.09345","created_at":"2026-07-13T01:19:04.095815+00:00"},{"alias_kind":"pith_short_12","alias_value":"SKK2DPKJJ5RL","created_at":"2026-07-13T01:19:04.095815+00:00"},{"alias_kind":"pith_short_16","alias_value":"SKK2DPKJJ5RLPDPP","created_at":"2026-07-13T01:19:04.095815+00:00"},{"alias_kind":"pith_short_8","alias_value":"SKK2DPKJ","created_at":"2026-07-13T01:19:04.095815+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/SKK2DPKJJ5RLPDPP2NLMYCBCXX","json":"https://pith.science/pith/SKK2DPKJJ5RLPDPP2NLMYCBCXX.json","graph_json":"https://pith.science/api/pith-number/SKK2DPKJJ5RLPDPP2NLMYCBCXX/graph.json","events_json":"https://pith.science/api/pith-number/SKK2DPKJJ5RLPDPP2NLMYCBCXX/events.json","paper":"https://pith.science/paper/SKK2DPKJ"},"agent_actions":{"view_html":"https://pith.science/pith/SKK2DPKJJ5RLPDPP2NLMYCBCXX","download_json":"https://pith.science/pith/SKK2DPKJJ5RLPDPP2NLMYCBCXX.json","view_paper":"https://pith.science/paper/SKK2DPKJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.09345&json=true","fetch_graph":"https://pith.science/api/pith-number/SKK2DPKJJ5RLPDPP2NLMYCBCXX/graph.json","fetch_events":"https://pith.science/api/pith-number/SKK2DPKJJ5RLPDPP2NLMYCBCXX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SKK2DPKJJ5RLPDPP2NLMYCBCXX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SKK2DPKJJ5RLPDPP2NLMYCBCXX/action/storage_attestation","attest_author":"https://pith.science/pith/SKK2DPKJJ5RLPDPP2NLMYCBCXX/action/author_attestation","sign_citation":"https://pith.science/pith/SKK2DPKJJ5RLPDPP2NLMYCBCXX/action/citation_signature","submit_replication":"https://pith.science/pith/SKK2DPKJJ5RLPDPP2NLMYCBCXX/action/replication_record"}},"created_at":"2026-07-13T01:19:04.095815+00:00","updated_at":"2026-07-13T01:19:04.095815+00:00"}