{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:OTM3EZY5N7EYNPTCVYM7B7HIGW","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":"d1f88929215018b48d39d93a60518d70451ec73e1e3a58933535d3f8f49593dd","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-09T08:01:55Z","title_canon_sha256":"c683d57e488c4a58a7b52d74d0f499b2347c7a4f4d3b54835b628187dbf05739"},"schema_version":"1.0","source":{"id":"2607.08203","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.08203","created_at":"2026-07-10T01:19:29Z"},{"alias_kind":"arxiv_version","alias_value":"2607.08203v1","created_at":"2026-07-10T01:19:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.08203","created_at":"2026-07-10T01:19:29Z"},{"alias_kind":"pith_short_12","alias_value":"OTM3EZY5N7EY","created_at":"2026-07-10T01:19:29Z"},{"alias_kind":"pith_short_16","alias_value":"OTM3EZY5N7EYNPTC","created_at":"2026-07-10T01:19:29Z"},{"alias_kind":"pith_short_8","alias_value":"OTM3EZY5","created_at":"2026-07-10T01:19:29Z"}],"graph_snapshots":[{"event_id":"sha256:a91604c2776b8cf06d8dd4727bf882cddb915edf467ac9259fe2c8708dcc78cc","target":"graph","created_at":"2026-07-10T01:19:29Z","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/2607.08203/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Progress in colonoscopy polyp segmentation is routinely reported through leaderboard comparisons on a small set of public benchmarks. We argue that this apparent progress is difficult to verify: a systematic audit of \\textbf{27 papers} published between 2015 and 2026 reveals three structural problems in how the community evaluates models. \\textbf{First}, 25 of 27 papers \\textit{omit the Hausdorff distance}. Hausdorff distance is a boundary-accuracy metric with direct clinical relevance for detecting flat or small polyps, and is a standard in radiotherapy segmentation. \\textbf{Second}, at least","authors_text":"Aisha Urooj, Neelu Madan, Zain Ul Abdien","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-09T08:01:55Z","title":"Metrics or Mirage? An Audit of Evaluation Inconsistencies in Colonoscopy Polyp Segmentation Benchmarks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.08203","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":"sha256:19afd1339208be1992a99517a2066860a712b4634339381e8c94e2de2f8cee54","target":"record","created_at":"2026-07-10T01:19:29Z","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":"d1f88929215018b48d39d93a60518d70451ec73e1e3a58933535d3f8f49593dd","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-09T08:01:55Z","title_canon_sha256":"c683d57e488c4a58a7b52d74d0f499b2347c7a4f4d3b54835b628187dbf05739"},"schema_version":"1.0","source":{"id":"2607.08203","kind":"arxiv","version":1}},"canonical_sha256":"74d9b2671d6fc986be62ae19f0fce83590a46a790dd0c491847765012267ac41","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"74d9b2671d6fc986be62ae19f0fce83590a46a790dd0c491847765012267ac41","first_computed_at":"2026-07-10T01:19:29.638579Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-10T01:19:29.638579Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"JHzIt2CBKjLpNeigbUrcE/JpFbnLL3WoT0OhDFO0lBP1WXF6hYhCrex+7d/62l3JNLNcjQQNfT5/K5CS+4ZPCA==","signature_status":"signed_v1","signed_at":"2026-07-10T01:19:29.638980Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.08203","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:19afd1339208be1992a99517a2066860a712b4634339381e8c94e2de2f8cee54","sha256:a91604c2776b8cf06d8dd4727bf882cddb915edf467ac9259fe2c8708dcc78cc"],"state_sha256":"652194536b8cebb8f2d8dfd5b2bf083bb5fd42c782b84e4e8b0923374d1c5a70"}