{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:MKGVQJXHKFDE3EWRDM3GGQAQMZ","short_pith_number":"pith:MKGVQJXH","schema_version":"1.0","canonical_sha256":"628d5826e751464d92d11b366340106641d83562e40df34beb5ec09bb9ae63d3","source":{"kind":"arxiv","id":"2606.26041","version":1},"attestation_state":"computed","paper":{"title":"How Robust is OCR-Reasoning? Evaluating OCR-Reasoning Robustness of Vision-Language Models under Visual Perturbations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Yi Chang, Yuan Wu, Yuxing Cheng","submitted_at":"2026-06-24T17:15:42Z","abstract_excerpt":"Vision-language models (VLMs) have achieved strong performance on OCR-based benchmarks and increasingly focused on text-rich understanding, but their robustness under controlled visual degradation remains insufficiently understood. This gap is critical for OCR reasoning, where visual corruption can induce OCR errors and structural distortions, thereby introducing uncertainty into the reasoning task. To systematically study this problem, we introduce OCR-Robust, a benchmark designed for evaluating OCR reasoning robustness under visual perturbations. It contains 812 samples across two complement"},"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":"2606.26041","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-24T17:15:42Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"4620f4b8c49e2798eec316cd2a29e613afe51665e48f4b2e075895a67a83047b","abstract_canon_sha256":"c10b2ee936e61ea6faffd59489feec095508f189fb6225aa00e713931324ccbf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-25T01:18:46.659979Z","signature_b64":"G9GyFFvpVZjCVuIHunx8XiIPPokoV5Tp17hO3WXbMg0uLFM2iGoN8OH/qworqfp3bZfbvz1GzZXDNPcHAq9qCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"628d5826e751464d92d11b366340106641d83562e40df34beb5ec09bb9ae63d3","last_reissued_at":"2026-06-25T01:18:46.659579Z","signature_status":"signed_v1","first_computed_at":"2026-06-25T01:18:46.659579Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"How Robust is OCR-Reasoning? Evaluating OCR-Reasoning Robustness of Vision-Language Models under Visual Perturbations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Yi Chang, Yuan Wu, Yuxing Cheng","submitted_at":"2026-06-24T17:15:42Z","abstract_excerpt":"Vision-language models (VLMs) have achieved strong performance on OCR-based benchmarks and increasingly focused on text-rich understanding, but their robustness under controlled visual degradation remains insufficiently understood. This gap is critical for OCR reasoning, where visual corruption can induce OCR errors and structural distortions, thereby introducing uncertainty into the reasoning task. To systematically study this problem, we introduce OCR-Robust, a benchmark designed for evaluating OCR reasoning robustness under visual perturbations. It contains 812 samples across two complement"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.26041","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/2606.26041/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":"2606.26041","created_at":"2026-06-25T01:18:46.659644+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.26041v1","created_at":"2026-06-25T01:18:46.659644+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.26041","created_at":"2026-06-25T01:18:46.659644+00:00"},{"alias_kind":"pith_short_12","alias_value":"MKGVQJXHKFDE","created_at":"2026-06-25T01:18:46.659644+00:00"},{"alias_kind":"pith_short_16","alias_value":"MKGVQJXHKFDE3EWR","created_at":"2026-06-25T01:18:46.659644+00:00"},{"alias_kind":"pith_short_8","alias_value":"MKGVQJXH","created_at":"2026-06-25T01:18:46.659644+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/MKGVQJXHKFDE3EWRDM3GGQAQMZ","json":"https://pith.science/pith/MKGVQJXHKFDE3EWRDM3GGQAQMZ.json","graph_json":"https://pith.science/api/pith-number/MKGVQJXHKFDE3EWRDM3GGQAQMZ/graph.json","events_json":"https://pith.science/api/pith-number/MKGVQJXHKFDE3EWRDM3GGQAQMZ/events.json","paper":"https://pith.science/paper/MKGVQJXH"},"agent_actions":{"view_html":"https://pith.science/pith/MKGVQJXHKFDE3EWRDM3GGQAQMZ","download_json":"https://pith.science/pith/MKGVQJXHKFDE3EWRDM3GGQAQMZ.json","view_paper":"https://pith.science/paper/MKGVQJXH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.26041&json=true","fetch_graph":"https://pith.science/api/pith-number/MKGVQJXHKFDE3EWRDM3GGQAQMZ/graph.json","fetch_events":"https://pith.science/api/pith-number/MKGVQJXHKFDE3EWRDM3GGQAQMZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MKGVQJXHKFDE3EWRDM3GGQAQMZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MKGVQJXHKFDE3EWRDM3GGQAQMZ/action/storage_attestation","attest_author":"https://pith.science/pith/MKGVQJXHKFDE3EWRDM3GGQAQMZ/action/author_attestation","sign_citation":"https://pith.science/pith/MKGVQJXHKFDE3EWRDM3GGQAQMZ/action/citation_signature","submit_replication":"https://pith.science/pith/MKGVQJXHKFDE3EWRDM3GGQAQMZ/action/replication_record"}},"created_at":"2026-06-25T01:18:46.659644+00:00","updated_at":"2026-06-25T01:18:46.659644+00:00"}