{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:A5YDA6LL5OKCZEHTCDKMW3OSNS","short_pith_number":"pith:A5YDA6LL","schema_version":"1.0","canonical_sha256":"077030796beb942c90f310d4cb6dd26c8649e3400cd2a8277626e068ad493104","source":{"kind":"arxiv","id":"2505.22063","version":1},"attestation_state":"computed","paper":{"title":"Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.SD","authors_text":"Bingshen Mu, Chengyou Wang, Danming Xie, Hai Li, Junhui Liu, Lei Xie, Mingchen Shao, Xinfa Zhu, Ying Yan","submitted_at":"2025-05-28T07:39:25Z","abstract_excerpt":"Despite remarkable achievements, automatic speech recognition (ASR) in low-resource scenarios still faces two challenges: high-quality data scarcity and high computational demands. This paper proposes EThai-ASR, the first to apply large language models (LLMs) to Thai ASR and create an efficient LLM-based ASR system. EThai-ASR comprises a speech encoder, a connection module and a Thai LLM decoder. To address the data scarcity and obtain a powerful speech encoder, EThai-ASR introduces a self-evolving data refinement strategy to refine weak labels, yielding an enhanced speech encoder. Moreover, w"},"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":"2505.22063","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SD","submitted_at":"2025-05-28T07:39:25Z","cross_cats_sorted":["eess.AS"],"title_canon_sha256":"322133dddfa8d3809405f835f266aa1e57d7f7ef6576a06336974508ebbc45ea","abstract_canon_sha256":"62e14480cc673baf27489fbf4418dc10c52a507c51022f7b3eee455e8fdb90f8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:11:09.041113Z","signature_b64":"m7aoIB0A8Y+3evgrPwHV8zOsD3iuASRS6eHb36Af0Lpgp9N0P/r4k82726EvLYoD9muvvNaMfiEnlGe1EMc1DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"077030796beb942c90f310d4cb6dd26c8649e3400cd2a8277626e068ad493104","last_reissued_at":"2026-07-05T11:11:09.040581Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:11:09.040581Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.SD","authors_text":"Bingshen Mu, Chengyou Wang, Danming Xie, Hai Li, Junhui Liu, Lei Xie, Mingchen Shao, Xinfa Zhu, Ying Yan","submitted_at":"2025-05-28T07:39:25Z","abstract_excerpt":"Despite remarkable achievements, automatic speech recognition (ASR) in low-resource scenarios still faces two challenges: high-quality data scarcity and high computational demands. This paper proposes EThai-ASR, the first to apply large language models (LLMs) to Thai ASR and create an efficient LLM-based ASR system. EThai-ASR comprises a speech encoder, a connection module and a Thai LLM decoder. To address the data scarcity and obtain a powerful speech encoder, EThai-ASR introduces a self-evolving data refinement strategy to refine weak labels, yielding an enhanced speech encoder. Moreover, w"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.22063","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/2505.22063/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":"2505.22063","created_at":"2026-07-05T11:11:09.040648+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.22063v1","created_at":"2026-07-05T11:11:09.040648+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.22063","created_at":"2026-07-05T11:11:09.040648+00:00"},{"alias_kind":"pith_short_12","alias_value":"A5YDA6LL5OKC","created_at":"2026-07-05T11:11:09.040648+00:00"},{"alias_kind":"pith_short_16","alias_value":"A5YDA6LL5OKCZEHT","created_at":"2026-07-05T11:11:09.040648+00:00"},{"alias_kind":"pith_short_8","alias_value":"A5YDA6LL","created_at":"2026-07-05T11:11:09.040648+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2509.14804","citing_title":"Towards Building Speech Large Language Models for Multitask Understanding in Low-Resource Languages","ref_index":25,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/A5YDA6LL5OKCZEHTCDKMW3OSNS","json":"https://pith.science/pith/A5YDA6LL5OKCZEHTCDKMW3OSNS.json","graph_json":"https://pith.science/api/pith-number/A5YDA6LL5OKCZEHTCDKMW3OSNS/graph.json","events_json":"https://pith.science/api/pith-number/A5YDA6LL5OKCZEHTCDKMW3OSNS/events.json","paper":"https://pith.science/paper/A5YDA6LL"},"agent_actions":{"view_html":"https://pith.science/pith/A5YDA6LL5OKCZEHTCDKMW3OSNS","download_json":"https://pith.science/pith/A5YDA6LL5OKCZEHTCDKMW3OSNS.json","view_paper":"https://pith.science/paper/A5YDA6LL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.22063&json=true","fetch_graph":"https://pith.science/api/pith-number/A5YDA6LL5OKCZEHTCDKMW3OSNS/graph.json","fetch_events":"https://pith.science/api/pith-number/A5YDA6LL5OKCZEHTCDKMW3OSNS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/A5YDA6LL5OKCZEHTCDKMW3OSNS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/A5YDA6LL5OKCZEHTCDKMW3OSNS/action/storage_attestation","attest_author":"https://pith.science/pith/A5YDA6LL5OKCZEHTCDKMW3OSNS/action/author_attestation","sign_citation":"https://pith.science/pith/A5YDA6LL5OKCZEHTCDKMW3OSNS/action/citation_signature","submit_replication":"https://pith.science/pith/A5YDA6LL5OKCZEHTCDKMW3OSNS/action/replication_record"}},"created_at":"2026-07-05T11:11:09.040648+00:00","updated_at":"2026-07-05T11:11:09.040648+00:00"}