{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:R6IC4VPSWA6PEUC6TMNBQXBEYP","short_pith_number":"pith:R6IC4VPS","schema_version":"1.0","canonical_sha256":"8f902e55f2b03cf2505e9b1a185c24c3ce2bd2cf36a962b361cba6242515181f","source":{"kind":"arxiv","id":"2605.27750","version":1},"attestation_state":"computed","paper":{"title":"Reading or Guessing? Visual Grounding Failures of Vision-Language Models for OCR in Ancient Greek Editions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.DL"],"primary_cat":"cs.CL","authors_text":"Antonia Karamolegkou, Beno\\^it Sagot, Nicolas Angleraud, Thibault Cl\\'erice","submitted_at":"2026-05-26T22:57:01Z","abstract_excerpt":"Recent work has shown that Vision-Language Models (VLMs) used for optical character recognition (OCR) can generate plausible but visually unsupported text, suggesting reliance on language priors. Comparing open-weight VLMs with traditional OCR baselines on low-resource Ancient Greek critical editions, we show that VLM errors often remain fluent even when wrong, producing plausible Greek substitutions where traditional engines produce local recognition noise. To analyze visual evidence during decoding, we introduce controlled image perturbations and token-level grounding measures based on condi"},"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":"2605.27750","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-26T22:57:01Z","cross_cats_sorted":["cs.AI","cs.CV","cs.DL"],"title_canon_sha256":"b0799685d617b499e725027486d3e51941d3548e55bd4291e1a7c389ec49d6d4","abstract_canon_sha256":"7fc1bf245ee488d6b33c7524e8b09a559d2376d96d966e2eb12814b39a667d54"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-28T01:04:47.883809Z","signature_b64":"t3yu6gy9w7Qbo8bBd4+yq03RJnjdSXnkoKXEGVz7a5rbVARK7NJpiwhjHLNUGOI6k2dy24WmXBvi+vW8gJgzAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8f902e55f2b03cf2505e9b1a185c24c3ce2bd2cf36a962b361cba6242515181f","last_reissued_at":"2026-05-28T01:04:47.883301Z","signature_status":"signed_v1","first_computed_at":"2026-05-28T01:04:47.883301Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reading or Guessing? Visual Grounding Failures of Vision-Language Models for OCR in Ancient Greek Editions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.DL"],"primary_cat":"cs.CL","authors_text":"Antonia Karamolegkou, Beno\\^it Sagot, Nicolas Angleraud, Thibault Cl\\'erice","submitted_at":"2026-05-26T22:57:01Z","abstract_excerpt":"Recent work has shown that Vision-Language Models (VLMs) used for optical character recognition (OCR) can generate plausible but visually unsupported text, suggesting reliance on language priors. Comparing open-weight VLMs with traditional OCR baselines on low-resource Ancient Greek critical editions, we show that VLM errors often remain fluent even when wrong, producing plausible Greek substitutions where traditional engines produce local recognition noise. To analyze visual evidence during decoding, we introduce controlled image perturbations and token-level grounding measures based on condi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.27750","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/2605.27750/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":"2605.27750","created_at":"2026-05-28T01:04:47.883380+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.27750v1","created_at":"2026-05-28T01:04:47.883380+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.27750","created_at":"2026-05-28T01:04:47.883380+00:00"},{"alias_kind":"pith_short_12","alias_value":"R6IC4VPSWA6P","created_at":"2026-05-28T01:04:47.883380+00:00"},{"alias_kind":"pith_short_16","alias_value":"R6IC4VPSWA6PEUC6","created_at":"2026-05-28T01:04:47.883380+00:00"},{"alias_kind":"pith_short_8","alias_value":"R6IC4VPS","created_at":"2026-05-28T01:04:47.883380+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/R6IC4VPSWA6PEUC6TMNBQXBEYP","json":"https://pith.science/pith/R6IC4VPSWA6PEUC6TMNBQXBEYP.json","graph_json":"https://pith.science/api/pith-number/R6IC4VPSWA6PEUC6TMNBQXBEYP/graph.json","events_json":"https://pith.science/api/pith-number/R6IC4VPSWA6PEUC6TMNBQXBEYP/events.json","paper":"https://pith.science/paper/R6IC4VPS"},"agent_actions":{"view_html":"https://pith.science/pith/R6IC4VPSWA6PEUC6TMNBQXBEYP","download_json":"https://pith.science/pith/R6IC4VPSWA6PEUC6TMNBQXBEYP.json","view_paper":"https://pith.science/paper/R6IC4VPS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.27750&json=true","fetch_graph":"https://pith.science/api/pith-number/R6IC4VPSWA6PEUC6TMNBQXBEYP/graph.json","fetch_events":"https://pith.science/api/pith-number/R6IC4VPSWA6PEUC6TMNBQXBEYP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/R6IC4VPSWA6PEUC6TMNBQXBEYP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/R6IC4VPSWA6PEUC6TMNBQXBEYP/action/storage_attestation","attest_author":"https://pith.science/pith/R6IC4VPSWA6PEUC6TMNBQXBEYP/action/author_attestation","sign_citation":"https://pith.science/pith/R6IC4VPSWA6PEUC6TMNBQXBEYP/action/citation_signature","submit_replication":"https://pith.science/pith/R6IC4VPSWA6PEUC6TMNBQXBEYP/action/replication_record"}},"created_at":"2026-05-28T01:04:47.883380+00:00","updated_at":"2026-05-28T01:04:47.883380+00:00"}