{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2020:3VJNXMM47PTHG2SFTFVSR4MWNB","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"25b7199e0f3cc792fbfebfa16ca19f96062c8386669a700c0624a1eb692aafd8","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-07-10T17:25:11Z","title_canon_sha256":"564b695f38a6813333fed8c82fd0fd6864c985f7a0a346197073c2e9cd2aa7ea"},"schema_version":"1.0","source":{"id":"2007.05499","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2007.05499","created_at":"2026-07-05T01:17:47Z"},{"alias_kind":"arxiv_version","alias_value":"2007.05499v1","created_at":"2026-07-05T01:17:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2007.05499","created_at":"2026-07-05T01:17:47Z"},{"alias_kind":"pith_short_12","alias_value":"3VJNXMM47PTH","created_at":"2026-07-05T01:17:47Z"},{"alias_kind":"pith_short_16","alias_value":"3VJNXMM47PTHG2SF","created_at":"2026-07-05T01:17:47Z"},{"alias_kind":"pith_short_8","alias_value":"3VJNXMM4","created_at":"2026-07-05T01:17:47Z"}],"graph_snapshots":[{"event_id":"sha256:f4060f1291791ce11238cdf08a245ba740ff5ec9de41f1d13fb2fbc63a9bc868","target":"graph","created_at":"2026-07-05T01:17:47Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2007.05499/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Building and maintaining high-quality test sets remains a laborious and expensive task. As a result, test sets in the real world are often not properly kept up to date and drift from the production traffic they are supposed to represent. The frequency and severity of this drift raises serious concerns over the value of manually labeled test sets in the QA process. This paper proposes a simple but effective technique that drastically reduces the effort needed to construct and maintain a high-quality test set (reducing labeling effort by 80-100% across a range of practical scenarios). This resul","authors_text":"Anupama Murthi, Begum Taskazan, Benjamin Elder, Ganesh Venkataraman, Jiri Navratil, Matthew Arnold","cross_cats":["stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-07-10T17:25:11Z","title":"Not Your Grandfathers Test Set: Reducing Labeling Effort for Testing"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2007.05499","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:c571d39f820f194b4fd51d6a29bc7f6d6e1c2ad91229efdfd091603813261135","target":"record","created_at":"2026-07-05T01:17:47Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"25b7199e0f3cc792fbfebfa16ca19f96062c8386669a700c0624a1eb692aafd8","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-07-10T17:25:11Z","title_canon_sha256":"564b695f38a6813333fed8c82fd0fd6864c985f7a0a346197073c2e9cd2aa7ea"},"schema_version":"1.0","source":{"id":"2007.05499","kind":"arxiv","version":1}},"canonical_sha256":"dd52dbb19cfbe6736a45996b28f1966866f53d301db1d528294d32c1a33a9997","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"dd52dbb19cfbe6736a45996b28f1966866f53d301db1d528294d32c1a33a9997","first_computed_at":"2026-07-05T01:17:47.510708Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T01:17:47.510708Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"GW+lBNv6+lrNPeebSX2/7q/25kyTnkOxEAbA93ZGyaKyXjxkTiQ++srwtB570zaOkunHuSZTB23fCpBA2gFmCw==","signature_status":"signed_v1","signed_at":"2026-07-05T01:17:47.511131Z","signed_message":"canonical_sha256_bytes"},"source_id":"2007.05499","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c571d39f820f194b4fd51d6a29bc7f6d6e1c2ad91229efdfd091603813261135","sha256:f4060f1291791ce11238cdf08a245ba740ff5ec9de41f1d13fb2fbc63a9bc868"],"state_sha256":"7e7413540f8fbaafad20fdbf17dc1bc8150ff4aa097cac2b5a0f6a0f39a89421"}