{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:NQYXZCKCANGSIAIJLWC56WWKO7","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"6e4405473836b2bcd83923320527c4db4c08ab6800ae7e9772add5a45cc45617","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SD","submitted_at":"2026-03-05T12:07:19Z","title_canon_sha256":"2fce4f32fe14928925971bf796b9abd91738de69487a5374ca40333c528c3b20"},"schema_version":"1.0","source":{"id":"2603.05094","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2603.05094","created_at":"2026-05-18T03:09:23Z"},{"alias_kind":"arxiv_version","alias_value":"2603.05094v3","created_at":"2026-05-18T03:09:23Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.05094","created_at":"2026-05-18T03:09:23Z"},{"alias_kind":"pith_short_12","alias_value":"NQYXZCKCANGS","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"NQYXZCKCANGSIAIJ","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"NQYXZCKC","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:518262d54d00fe02ad64c68d1b8562c7cea692b5a082f27c3050bd51d83b85ef","target":"graph","created_at":"2026-05-18T03:09:23Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"On the TAU Benchmark, Tai-LALM reaches 49.1% accuracy, marking a 6.5% absolute improvement over the zero-shot baseline (42.6% with ASR text conditioning). This confirms that integrating regional corpora with rigorous curation and dynamic arbitration significantly enhances LALM performance on localized speech."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the Verify-Generate-Critique protocol combined with Dual-ASR validation produces genuinely higher-fidelity instruction pairs that causally drive the observed benchmark gain rather than other unstated differences in training or evaluation."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"TW-Sound580K dataset plus Tai-LALM model with dynamic Dual-ASR arbitration lifts localized Taiwanese audio-language accuracy to 49.1% on the TAU benchmark."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"A verification-curated Taiwanese audio-text dataset and dynamic arbitration strategy lifts audio-language model accuracy on localized speech from 42.6 to 49.1 percent."}],"snapshot_sha256":"7ccd80476394bdec8bf5cb7d768ec5865b0577f92801ffbee3586ac984f322d4"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"ce8cde31f26c05a060d8c8f3e25dbaf7445b98f2eedbd1df5f91db579f42d99d"},"paper":{"abstract_excerpt":"Large Audio-Language Models (LALMs) typically struggle with localized dialectal prosody due to the scarcity of specialized corpora. We present TW-Sound580K, a Taiwanese audio-text instruction dataset developed through a Verify-Generate-Critique (VGC) protocol. This pipeline leverages Dual-ASR validation to filter 522K raw clips, subsequently expanding them into 580,000 high-fidelity instruction pairs using a teacher model. The dataset's utility is demonstrated through Tai-LALM, which fine-tunes a DeSTA 2.5-Audio-initialized backbone and incorporates a dynamic Dual-ASR Arbitration strategy to o","authors_text":"Hao-Hui Xie, Ho-Lam Chung, Hung-yi Lee, Ke-Han Lu, Wenze Ren, Xie Chen, Yi-Cheng Lin","cross_cats":[],"headline":"A verification-curated Taiwanese audio-text dataset and dynamic arbitration strategy lifts audio-language model accuracy on localized speech from 42.6 to 49.1 percent.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SD","submitted_at":"2026-03-05T12:07:19Z","title":"TW-Sound580K: A Regional Audio-Text Dataset with Verification-Guided Curation for Localized Audio-Language Modeling"},"references":{"count":44,"internal_anchors":6,"resolved_work":44,"sample":[{"cited_arxiv_id":"2603.05094","doi":"","is_internal_anchor":true,"ref_index":1,"title":"TW-Sound580K: A Regional Audio-Text Dataset with Verification-Guided Curation for Localized Audio-Language Modeling","work_id":"6bb3cbea-64cd-42fe-9dce-42095e900ba7","year":2026},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Background Foundational corpora like AudioSet [11] and LibriSpeech","work_id":"b257d624-c32d-4720-9320-0f2e20fc8f86","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"acoustic long-tail","work_id":"fd705707-c127-46ca-bfed-6a6142fe9067","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"Methodology To bridge the localization gap, our data-centric pipeline is struc- tured into four key stages: (I) Dataset Construction, (II) Train- ing Data Generation, (III) the multimodal Training Pro","work_id":"9b9c31f1-05fe-4f1d-b5e4-34193e1f2072","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"To preserve speech-free soundmarks, clips where both ASRs yield empty outputs bypass the text check","work_id":"f0d7d77f-88ee-4324-b76d-6d1595bcbf34","year":null}],"snapshot_sha256":"812665e115ada450767d90a34e3ccb747a8b2d073e2b60e42baa072addb7034c"},"source":{"id":"2603.05094","kind":"arxiv","version":3},"verdict":{"created_at":"2026-05-15T15:36:28.639447Z","id":"5a8d4b94-47b6-43ca-9c3f-1bd09542495f","model_set":{"reader":"grok-4.3"},"one_line_summary":"TW-Sound580K dataset plus Tai-LALM model with dynamic Dual-ASR arbitration lifts localized Taiwanese audio-language accuracy to 49.1% on the TAU benchmark.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"A verification-curated Taiwanese audio-text dataset and dynamic arbitration strategy lifts audio-language model accuracy on localized speech from 42.6 to 49.1 percent.","strongest_claim":"On the TAU Benchmark, Tai-LALM reaches 49.1% accuracy, marking a 6.5% absolute improvement over the zero-shot baseline (42.6% with ASR text conditioning). This confirms that integrating regional corpora with rigorous curation and dynamic arbitration significantly enhances LALM performance on localized speech.","weakest_assumption":"That the Verify-Generate-Critique protocol combined with Dual-ASR validation produces genuinely higher-fidelity instruction pairs that causally drive the observed benchmark gain rather than other unstated differences in training or evaluation."}},"verdict_id":"5a8d4b94-47b6-43ca-9c3f-1bd09542495f"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:ffbe718e479648c337f954581639775fcb80d42759a277685085cd28aeb44014","target":"record","created_at":"2026-05-18T03:09:23Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"6e4405473836b2bcd83923320527c4db4c08ab6800ae7e9772add5a45cc45617","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SD","submitted_at":"2026-03-05T12:07:19Z","title_canon_sha256":"2fce4f32fe14928925971bf796b9abd91738de69487a5374ca40333c528c3b20"},"schema_version":"1.0","source":{"id":"2603.05094","kind":"arxiv","version":3}},"canonical_sha256":"6c317c8942034d2401095d85df5aca77ef6b71410b43bf4606c95b3955ab0bae","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6c317c8942034d2401095d85df5aca77ef6b71410b43bf4606c95b3955ab0bae","first_computed_at":"2026-05-18T03:09:23.120726Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T03:09:23.120726Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Z1GPfPrH/Qy+E370JJp3n0BjlZATvw8hb4FwvL1Cz+is9M5YQTEXn/ha+437/m4YzM+Ddrq7VBbrKzPXA3B9Cg==","signature_status":"signed_v1","signed_at":"2026-05-18T03:09:23.121885Z","signed_message":"canonical_sha256_bytes"},"source_id":"2603.05094","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ffbe718e479648c337f954581639775fcb80d42759a277685085cd28aeb44014","sha256:518262d54d00fe02ad64c68d1b8562c7cea692b5a082f27c3050bd51d83b85ef"],"state_sha256":"475a424279eaf6fe9a4512e35ff97857cadcfb9c1c2b8afdc28c6abd3930de35"}