{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:NZAMT27ZX7W4NJZWCL76IHXQWU","short_pith_number":"pith:NZAMT27Z","schema_version":"1.0","canonical_sha256":"6e40c9ebf9bfedc6a73612ffe41ef0b50e2fa3a55a71d9db9e690ac39d6141a4","source":{"kind":"arxiv","id":"2104.08211","version":3},"attestation_state":"computed","paper":{"title":"Robust Open-Vocabulary Translation from Visual Text Representations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"David Etter, Elizabeth Salesky, Matt Post","submitted_at":"2021-04-16T16:37:13Z","abstract_excerpt":"Machine translation models have discrete vocabularies and commonly use subword segmentation techniques to achieve an 'open vocabulary.' This approach relies on consistent and correct underlying unicode sequences, and makes models susceptible to degradation from common types of noise and variation. Motivated by the robustness of human language processing, we propose the use of visual text representations, which dispense with a finite set of text embeddings in favor of continuous vocabularies created by processing visually rendered text with sliding windows. We show that models using visual text"},"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":"2104.08211","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-04-16T16:37:13Z","cross_cats_sorted":[],"title_canon_sha256":"d28f71f7055f22d777b64d5ec3f0e8f515b5118606311a4c973361735be2cc36","abstract_canon_sha256":"498a1b887699862399e96513b7bef8172504733555cd52e4a1a5ab4331bf090d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:39:33.976596Z","signature_b64":"uMWqka5jqXSPrhS1ZOakN+YJRNRhpid2DXqTl7AwRJb/4GDVujIr9UxlbdLeyBm9WSdvIKEmEPO86uohs9EWDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6e40c9ebf9bfedc6a73612ffe41ef0b50e2fa3a55a71d9db9e690ac39d6141a4","last_reissued_at":"2026-07-05T03:39:33.975968Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:39:33.975968Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Robust Open-Vocabulary Translation from Visual Text Representations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"David Etter, Elizabeth Salesky, Matt Post","submitted_at":"2021-04-16T16:37:13Z","abstract_excerpt":"Machine translation models have discrete vocabularies and commonly use subword segmentation techniques to achieve an 'open vocabulary.' This approach relies on consistent and correct underlying unicode sequences, and makes models susceptible to degradation from common types of noise and variation. Motivated by the robustness of human language processing, we propose the use of visual text representations, which dispense with a finite set of text embeddings in favor of continuous vocabularies created by processing visually rendered text with sliding windows. We show that models using visual text"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2104.08211","kind":"arxiv","version":3},"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/2104.08211/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":"2104.08211","created_at":"2026-07-05T03:39:33.976028+00:00"},{"alias_kind":"arxiv_version","alias_value":"2104.08211v3","created_at":"2026-07-05T03:39:33.976028+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2104.08211","created_at":"2026-07-05T03:39:33.976028+00:00"},{"alias_kind":"pith_short_12","alias_value":"NZAMT27ZX7W4","created_at":"2026-07-05T03:39:33.976028+00:00"},{"alias_kind":"pith_short_16","alias_value":"NZAMT27ZX7W4NJZW","created_at":"2026-07-05T03:39:33.976028+00:00"},{"alias_kind":"pith_short_8","alias_value":"NZAMT27Z","created_at":"2026-07-05T03:39:33.976028+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/NZAMT27ZX7W4NJZWCL76IHXQWU","json":"https://pith.science/pith/NZAMT27ZX7W4NJZWCL76IHXQWU.json","graph_json":"https://pith.science/api/pith-number/NZAMT27ZX7W4NJZWCL76IHXQWU/graph.json","events_json":"https://pith.science/api/pith-number/NZAMT27ZX7W4NJZWCL76IHXQWU/events.json","paper":"https://pith.science/paper/NZAMT27Z"},"agent_actions":{"view_html":"https://pith.science/pith/NZAMT27ZX7W4NJZWCL76IHXQWU","download_json":"https://pith.science/pith/NZAMT27ZX7W4NJZWCL76IHXQWU.json","view_paper":"https://pith.science/paper/NZAMT27Z","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2104.08211&json=true","fetch_graph":"https://pith.science/api/pith-number/NZAMT27ZX7W4NJZWCL76IHXQWU/graph.json","fetch_events":"https://pith.science/api/pith-number/NZAMT27ZX7W4NJZWCL76IHXQWU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NZAMT27ZX7W4NJZWCL76IHXQWU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NZAMT27ZX7W4NJZWCL76IHXQWU/action/storage_attestation","attest_author":"https://pith.science/pith/NZAMT27ZX7W4NJZWCL76IHXQWU/action/author_attestation","sign_citation":"https://pith.science/pith/NZAMT27ZX7W4NJZWCL76IHXQWU/action/citation_signature","submit_replication":"https://pith.science/pith/NZAMT27ZX7W4NJZWCL76IHXQWU/action/replication_record"}},"created_at":"2026-07-05T03:39:33.976028+00:00","updated_at":"2026-07-05T03:39:33.976028+00:00"}