{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:GUIZUHKUCCVTINWQRG6CN5EPXH","short_pith_number":"pith:GUIZUHKU","schema_version":"1.0","canonical_sha256":"35119a1d5410ab3436d089bc26f48fb9ca98d103eef706b3dd3afd7273669217","source":{"kind":"arxiv","id":"2605.19711","version":1},"attestation_state":"computed","paper":{"title":"Can Large Language Models Reliably Correct Errors in Low-Resource ASR? A Contamination-Aware Case Study on West Frisian","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Martijn Wieling, Reihaneh Amooie, Rik van Noord, Wietse de Vries, Yun Hao","submitted_at":"2026-05-19T11:48:32Z","abstract_excerpt":"Automatic speech recognition (ASR) has improved substantially in recent years, yet performance remains limited for low-resource languages. Large language models (LLMs) have shown promise for improving ASR through generative error correction (GER), but their effectiveness in low-resource settings remains underexplored. In addition, it remains unclear to what extent data contamination influences the reported improvements in LLM-based GER. This study investigates LLM-based GER for low-resource Frisian. In addition to a public corpus, we construct and use a Frisian offline dataset with non-public "},"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":"2605.19711","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-19T11:48:32Z","cross_cats_sorted":[],"title_canon_sha256":"42d0b3e05adf4314c7c41e43f67f157fcb481fb2825e8f86185c7bece2ba893d","abstract_canon_sha256":"6d87e0bd189d9a97caeaa66143d16c8b8079d92d34bd68713bc5e11d5c71774b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T01:05:58.725925Z","signature_b64":"dDfAYH9cREJ5nKF5jWfVPBH00OJrlLQNI91Lvu2/WgQxM88EexWyw02N9Qmn4RLN5NT3Pyusl2J+ljNewds4Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"35119a1d5410ab3436d089bc26f48fb9ca98d103eef706b3dd3afd7273669217","last_reissued_at":"2026-05-20T01:05:58.725430Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T01:05:58.725430Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Can Large Language Models Reliably Correct Errors in Low-Resource ASR? A Contamination-Aware Case Study on West Frisian","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Martijn Wieling, Reihaneh Amooie, Rik van Noord, Wietse de Vries, Yun Hao","submitted_at":"2026-05-19T11:48:32Z","abstract_excerpt":"Automatic speech recognition (ASR) has improved substantially in recent years, yet performance remains limited for low-resource languages. Large language models (LLMs) have shown promise for improving ASR through generative error correction (GER), but their effectiveness in low-resource settings remains underexplored. In addition, it remains unclear to what extent data contamination influences the reported improvements in LLM-based GER. This study investigates LLM-based GER for low-resource Frisian. In addition to a public corpus, we construct and use a Frisian offline dataset with non-public "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.19711","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/2605.19711/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":"2605.19711","created_at":"2026-05-20T01:05:58.725495+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.19711v1","created_at":"2026-05-20T01:05:58.725495+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.19711","created_at":"2026-05-20T01:05:58.725495+00:00"},{"alias_kind":"pith_short_12","alias_value":"GUIZUHKUCCVT","created_at":"2026-05-20T01:05:58.725495+00:00"},{"alias_kind":"pith_short_16","alias_value":"GUIZUHKUCCVTINWQ","created_at":"2026-05-20T01:05:58.725495+00:00"},{"alias_kind":"pith_short_8","alias_value":"GUIZUHKU","created_at":"2026-05-20T01:05:58.725495+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.19711","citing_title":"Can Large Language Models Reliably Correct Errors in Low-Resource ASR? A Contamination-Aware Case Study on West Frisian","ref_index":2,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GUIZUHKUCCVTINWQRG6CN5EPXH","json":"https://pith.science/pith/GUIZUHKUCCVTINWQRG6CN5EPXH.json","graph_json":"https://pith.science/api/pith-number/GUIZUHKUCCVTINWQRG6CN5EPXH/graph.json","events_json":"https://pith.science/api/pith-number/GUIZUHKUCCVTINWQRG6CN5EPXH/events.json","paper":"https://pith.science/paper/GUIZUHKU"},"agent_actions":{"view_html":"https://pith.science/pith/GUIZUHKUCCVTINWQRG6CN5EPXH","download_json":"https://pith.science/pith/GUIZUHKUCCVTINWQRG6CN5EPXH.json","view_paper":"https://pith.science/paper/GUIZUHKU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.19711&json=true","fetch_graph":"https://pith.science/api/pith-number/GUIZUHKUCCVTINWQRG6CN5EPXH/graph.json","fetch_events":"https://pith.science/api/pith-number/GUIZUHKUCCVTINWQRG6CN5EPXH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GUIZUHKUCCVTINWQRG6CN5EPXH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GUIZUHKUCCVTINWQRG6CN5EPXH/action/storage_attestation","attest_author":"https://pith.science/pith/GUIZUHKUCCVTINWQRG6CN5EPXH/action/author_attestation","sign_citation":"https://pith.science/pith/GUIZUHKUCCVTINWQRG6CN5EPXH/action/citation_signature","submit_replication":"https://pith.science/pith/GUIZUHKUCCVTINWQRG6CN5EPXH/action/replication_record"}},"created_at":"2026-05-20T01:05:58.725495+00:00","updated_at":"2026-05-20T01:05:58.725495+00:00"}