{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:FHURX5KGAJVOJ5HZFJYOO2TDGM","short_pith_number":"pith:FHURX5KG","schema_version":"1.0","canonical_sha256":"29e91bf546026ae4f4f92a70e76a63330e289874cc5313304895717979bd1a88","source":{"kind":"arxiv","id":"2601.14251","version":2},"attestation_state":"computed","paper":{"title":"LightOnOCR: A 1B End-to-End Multilingual Vision-Language Model for State-of-the-Art OCR","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Adrien Cavaill\\`es, Baptiste Aubertin, Said Taghadouini","submitted_at":"2026-01-20T18:58:32Z","abstract_excerpt":"We present LightOnOCR-2-1B, a 1B-parameter end-to-end multilingual vision--language model that converts document images (e.g., PDFs) into clean, naturally ordered text without brittle OCR pipelines. Trained on a large-scale, high-quality distillation mix with strong coverage of scans, French documents, and scientific PDFs, LightOnOCR-2 achieves state-of-the-art results on OlmOCR-Bench while being 9$\\times$ smaller and substantially faster than prior best-performing models. We further extend the output format to predict normalized bounding boxes for embedded images, introducing localization dur"},"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":"2601.14251","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-01-20T18:58:32Z","cross_cats_sorted":[],"title_canon_sha256":"f7187e8d02b0c7f7865ff8176292b3a3467ba53aebeac99765434e0de06ac4b3","abstract_canon_sha256":"34c48cf2a97eb553ca0ea159b4ae326e8b9fc440edeca8cc4be1f2447dc556ad"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-01T01:17:45.952420Z","signature_b64":"LzjqPsi6fQS/yMcFeLoF5fcP6g05Rxb7O+N7/3zlValfQHMHehE+RTKvUynVo/B5ZmuRXQDoNRdquTd/WS7TDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"29e91bf546026ae4f4f92a70e76a63330e289874cc5313304895717979bd1a88","last_reissued_at":"2026-07-01T01:17:45.951944Z","signature_status":"signed_v1","first_computed_at":"2026-07-01T01:17:45.951944Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LightOnOCR: A 1B End-to-End Multilingual Vision-Language Model for State-of-the-Art OCR","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Adrien Cavaill\\`es, Baptiste Aubertin, Said Taghadouini","submitted_at":"2026-01-20T18:58:32Z","abstract_excerpt":"We present LightOnOCR-2-1B, a 1B-parameter end-to-end multilingual vision--language model that converts document images (e.g., PDFs) into clean, naturally ordered text without brittle OCR pipelines. Trained on a large-scale, high-quality distillation mix with strong coverage of scans, French documents, and scientific PDFs, LightOnOCR-2 achieves state-of-the-art results on OlmOCR-Bench while being 9$\\times$ smaller and substantially faster than prior best-performing models. We further extend the output format to predict normalized bounding boxes for embedded images, introducing localization dur"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2601.14251","kind":"arxiv","version":2},"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/2601.14251/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":"2601.14251","created_at":"2026-07-01T01:17:45.952001+00:00"},{"alias_kind":"arxiv_version","alias_value":"2601.14251v2","created_at":"2026-07-01T01:17:45.952001+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.14251","created_at":"2026-07-01T01:17:45.952001+00:00"},{"alias_kind":"pith_short_12","alias_value":"FHURX5KGAJVO","created_at":"2026-07-01T01:17:45.952001+00:00"},{"alias_kind":"pith_short_16","alias_value":"FHURX5KGAJVOJ5HZ","created_at":"2026-07-01T01:17:45.952001+00:00"},{"alias_kind":"pith_short_8","alias_value":"FHURX5KG","created_at":"2026-07-01T01:17:45.952001+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":9,"internal_anchor_count":9,"sample":[{"citing_arxiv_id":"2606.29378","citing_title":"Cross-Temporal Sinhala OCR: Page-Level Adaptation and Diachronic Analysis","ref_index":13,"is_internal_anchor":true},{"citing_arxiv_id":"2606.29905","citing_title":"StrucTab: A Structured Optimization Framework for Table Parsing","ref_index":43,"is_internal_anchor":true},{"citing_arxiv_id":"2605.26712","citing_title":"METATR: A Multilingual, Evolving Benchmark for Automatic Text Recognition","ref_index":29,"is_internal_anchor":true},{"citing_arxiv_id":"2605.27750","citing_title":"Reading or Guessing? Visual Grounding Failures of Vision-Language Models for OCR in Ancient Greek Editions","ref_index":3,"is_internal_anchor":true},{"citing_arxiv_id":"2606.04166","citing_title":"End-to-End Text Line Detection and Ordering","ref_index":27,"is_internal_anchor":true},{"citing_arxiv_id":"2605.00392","citing_title":"RTPrune: Reading-Twice Inspired Token Pruning for Efficient DeepSeek-OCR Inference","ref_index":14,"is_internal_anchor":true},{"citing_arxiv_id":"2605.00392","citing_title":"RTPrune: Reading-Twice Inspired Token Pruning for Efficient DeepSeek-OCR Inference","ref_index":14,"is_internal_anchor":true},{"citing_arxiv_id":"2604.22880","citing_title":"TexOCR: Advancing Document OCR Models for Compilable Page-to-LaTeX Reconstruction","ref_index":1,"is_internal_anchor":true},{"citing_arxiv_id":"2604.12978","citing_title":"GlotOCR Bench: OCR Models Still Struggle Beyond a Handful of Unicode Scripts","ref_index":54,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FHURX5KGAJVOJ5HZFJYOO2TDGM","json":"https://pith.science/pith/FHURX5KGAJVOJ5HZFJYOO2TDGM.json","graph_json":"https://pith.science/api/pith-number/FHURX5KGAJVOJ5HZFJYOO2TDGM/graph.json","events_json":"https://pith.science/api/pith-number/FHURX5KGAJVOJ5HZFJYOO2TDGM/events.json","paper":"https://pith.science/paper/FHURX5KG"},"agent_actions":{"view_html":"https://pith.science/pith/FHURX5KGAJVOJ5HZFJYOO2TDGM","download_json":"https://pith.science/pith/FHURX5KGAJVOJ5HZFJYOO2TDGM.json","view_paper":"https://pith.science/paper/FHURX5KG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2601.14251&json=true","fetch_graph":"https://pith.science/api/pith-number/FHURX5KGAJVOJ5HZFJYOO2TDGM/graph.json","fetch_events":"https://pith.science/api/pith-number/FHURX5KGAJVOJ5HZFJYOO2TDGM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FHURX5KGAJVOJ5HZFJYOO2TDGM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FHURX5KGAJVOJ5HZFJYOO2TDGM/action/storage_attestation","attest_author":"https://pith.science/pith/FHURX5KGAJVOJ5HZFJYOO2TDGM/action/author_attestation","sign_citation":"https://pith.science/pith/FHURX5KGAJVOJ5HZFJYOO2TDGM/action/citation_signature","submit_replication":"https://pith.science/pith/FHURX5KGAJVOJ5HZFJYOO2TDGM/action/replication_record"}},"created_at":"2026-07-01T01:17:45.952001+00:00","updated_at":"2026-07-01T01:17:45.952001+00:00"}