{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:CPX25CAY6JKFWC7E5LUTW5O2T7","short_pith_number":"pith:CPX25CAY","schema_version":"1.0","canonical_sha256":"13efae8818f2545b0be4eae93b75da9ff736b2474459ffcea60171a80ecd4ae8","source":{"kind":"arxiv","id":"2606.09334","version":1},"attestation_state":"computed","paper":{"title":"How Far Can Prompting Go for Minimal-Edit Ukrainian Grammatical Error Correction?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Artem Chernodub, Kateryna Karpo","submitted_at":"2026-06-08T11:00:56Z","abstract_excerpt":"Fine-tuned Large Language Models (LLMs) dominate in Ukrainian grammatical error correction (GEC), while API-accessed LLMs remain nearly untested on minimal-edit benchmarks. We evaluate 11 commercial LLMs from four providers and one open-source Ukrainian model on the UNLP 2023 GEC-only benchmark, comparing zero-shot, few-shot, minimal-edits, and LLM-assisted prompt optimization strategies. Our best configuration (Gemini 3.1-Pro) reaches F0.5=69.22, closing over 90% of the gap to fine-tuned SOTA (F0.5=73.14). For zero-shot prompts, only Claude models benefit from Ukrainian instructions. However,"},"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":"2606.09334","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-08T11:00:56Z","cross_cats_sorted":[],"title_canon_sha256":"39ae5b924f4dfa144c79f5849ced551f6afba581dae8fcfa10405fb8a5800133","abstract_canon_sha256":"8ec3a31d128c4dba82492d9bbf25d1d08be06042292b9016a2135eed32059b70"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-09T02:08:15.984716Z","signature_b64":"eHNLhjxepGPxWsHbOtPgPO1dAfxdJOhkb/23/er6mWtYKHpp8W2+WtWbtZMut5aPpq0ZLoOzjcPfuv+4S1YaDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"13efae8818f2545b0be4eae93b75da9ff736b2474459ffcea60171a80ecd4ae8","last_reissued_at":"2026-06-09T02:08:15.983971Z","signature_status":"signed_v1","first_computed_at":"2026-06-09T02:08:15.983971Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"How Far Can Prompting Go for Minimal-Edit Ukrainian Grammatical Error Correction?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Artem Chernodub, Kateryna Karpo","submitted_at":"2026-06-08T11:00:56Z","abstract_excerpt":"Fine-tuned Large Language Models (LLMs) dominate in Ukrainian grammatical error correction (GEC), while API-accessed LLMs remain nearly untested on minimal-edit benchmarks. We evaluate 11 commercial LLMs from four providers and one open-source Ukrainian model on the UNLP 2023 GEC-only benchmark, comparing zero-shot, few-shot, minimal-edits, and LLM-assisted prompt optimization strategies. Our best configuration (Gemini 3.1-Pro) reaches F0.5=69.22, closing over 90% of the gap to fine-tuned SOTA (F0.5=73.14). For zero-shot prompts, only Claude models benefit from Ukrainian instructions. However,"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.09334","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/2606.09334/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":"2606.09334","created_at":"2026-06-09T02:08:15.984084+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.09334v1","created_at":"2026-06-09T02:08:15.984084+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.09334","created_at":"2026-06-09T02:08:15.984084+00:00"},{"alias_kind":"pith_short_12","alias_value":"CPX25CAY6JKF","created_at":"2026-06-09T02:08:15.984084+00:00"},{"alias_kind":"pith_short_16","alias_value":"CPX25CAY6JKFWC7E","created_at":"2026-06-09T02:08:15.984084+00:00"},{"alias_kind":"pith_short_8","alias_value":"CPX25CAY","created_at":"2026-06-09T02:08:15.984084+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/CPX25CAY6JKFWC7E5LUTW5O2T7","json":"https://pith.science/pith/CPX25CAY6JKFWC7E5LUTW5O2T7.json","graph_json":"https://pith.science/api/pith-number/CPX25CAY6JKFWC7E5LUTW5O2T7/graph.json","events_json":"https://pith.science/api/pith-number/CPX25CAY6JKFWC7E5LUTW5O2T7/events.json","paper":"https://pith.science/paper/CPX25CAY"},"agent_actions":{"view_html":"https://pith.science/pith/CPX25CAY6JKFWC7E5LUTW5O2T7","download_json":"https://pith.science/pith/CPX25CAY6JKFWC7E5LUTW5O2T7.json","view_paper":"https://pith.science/paper/CPX25CAY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.09334&json=true","fetch_graph":"https://pith.science/api/pith-number/CPX25CAY6JKFWC7E5LUTW5O2T7/graph.json","fetch_events":"https://pith.science/api/pith-number/CPX25CAY6JKFWC7E5LUTW5O2T7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CPX25CAY6JKFWC7E5LUTW5O2T7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CPX25CAY6JKFWC7E5LUTW5O2T7/action/storage_attestation","attest_author":"https://pith.science/pith/CPX25CAY6JKFWC7E5LUTW5O2T7/action/author_attestation","sign_citation":"https://pith.science/pith/CPX25CAY6JKFWC7E5LUTW5O2T7/action/citation_signature","submit_replication":"https://pith.science/pith/CPX25CAY6JKFWC7E5LUTW5O2T7/action/replication_record"}},"created_at":"2026-06-09T02:08:15.984084+00:00","updated_at":"2026-06-09T02:08:15.984084+00:00"}