{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:B2Y4EEKPCFSZS4PE2X6OF7XYD7","short_pith_number":"pith:B2Y4EEKP","schema_version":"1.0","canonical_sha256":"0eb1c2114f11659971e4d5fce2fef81fcac9c78bc076bb51a41ce70eaf248a7c","source":{"kind":"arxiv","id":"2409.19878","version":3},"attestation_state":"computed","paper":{"title":"HDMoLE: Mixture of LoRA Experts with Hierarchical Routing and Dynamic Thresholds for Fine-Tuning LLM-based ASR Models","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.SD","authors_text":"Bingshen Mu, Kun Wei, Lei Xie, Qijie Shao, Yong Xu","submitted_at":"2024-09-30T02:23:31Z","abstract_excerpt":"Recent advancements in integrating Large Language Models (LLM) with automatic speech recognition (ASR) have performed remarkably in general domains. While supervised fine-tuning (SFT) of all model parameters is often employed to adapt pre-trained LLM-based ASR models to specific domains, it imposes high computational costs and notably reduces their performance in general domains. In this paper, we propose a novel parameter-efficient multi-domain fine-tuning method for adapting pre-trained LLM-based ASR models to multi-accent domains without catastrophic forgetting named \\textit{HDMoLE}, which "},"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":"2409.19878","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.SD","submitted_at":"2024-09-30T02:23:31Z","cross_cats_sorted":["eess.AS"],"title_canon_sha256":"f4389950624a294a35f070b20d6fe38eb1f1e28fc4da265759647ebd669edc50","abstract_canon_sha256":"3c81b8633715a8e9fe855ea62d645a7b6238dccdd8d4fcb14a75aa6e0d205e99"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:56:22.255203Z","signature_b64":"a2pWDBeFqC1cUkIHAeLdwNjMOePUxgrfAPIrFfMl89MtbcQh9PFQtWOAn5txF5fuMmOvcRerdgmCllPR121pAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0eb1c2114f11659971e4d5fce2fef81fcac9c78bc076bb51a41ce70eaf248a7c","last_reissued_at":"2026-07-05T09:56:22.254712Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:56:22.254712Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"HDMoLE: Mixture of LoRA Experts with Hierarchical Routing and Dynamic Thresholds for Fine-Tuning LLM-based ASR Models","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.SD","authors_text":"Bingshen Mu, Kun Wei, Lei Xie, Qijie Shao, Yong Xu","submitted_at":"2024-09-30T02:23:31Z","abstract_excerpt":"Recent advancements in integrating Large Language Models (LLM) with automatic speech recognition (ASR) have performed remarkably in general domains. While supervised fine-tuning (SFT) of all model parameters is often employed to adapt pre-trained LLM-based ASR models to specific domains, it imposes high computational costs and notably reduces their performance in general domains. In this paper, we propose a novel parameter-efficient multi-domain fine-tuning method for adapting pre-trained LLM-based ASR models to multi-accent domains without catastrophic forgetting named \\textit{HDMoLE}, which "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.19878","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/2409.19878/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":"2409.19878","created_at":"2026-07-05T09:56:22.254773+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.19878v3","created_at":"2026-07-05T09:56:22.254773+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.19878","created_at":"2026-07-05T09:56:22.254773+00:00"},{"alias_kind":"pith_short_12","alias_value":"B2Y4EEKPCFSZ","created_at":"2026-07-05T09:56:22.254773+00:00"},{"alias_kind":"pith_short_16","alias_value":"B2Y4EEKPCFSZS4PE","created_at":"2026-07-05T09:56:22.254773+00:00"},{"alias_kind":"pith_short_8","alias_value":"B2Y4EEKP","created_at":"2026-07-05T09:56:22.254773+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.04167","citing_title":"Bench-CoE: a Framework for Collaboration of Experts from Benchmark","ref_index":19,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/B2Y4EEKPCFSZS4PE2X6OF7XYD7","json":"https://pith.science/pith/B2Y4EEKPCFSZS4PE2X6OF7XYD7.json","graph_json":"https://pith.science/api/pith-number/B2Y4EEKPCFSZS4PE2X6OF7XYD7/graph.json","events_json":"https://pith.science/api/pith-number/B2Y4EEKPCFSZS4PE2X6OF7XYD7/events.json","paper":"https://pith.science/paper/B2Y4EEKP"},"agent_actions":{"view_html":"https://pith.science/pith/B2Y4EEKPCFSZS4PE2X6OF7XYD7","download_json":"https://pith.science/pith/B2Y4EEKPCFSZS4PE2X6OF7XYD7.json","view_paper":"https://pith.science/paper/B2Y4EEKP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.19878&json=true","fetch_graph":"https://pith.science/api/pith-number/B2Y4EEKPCFSZS4PE2X6OF7XYD7/graph.json","fetch_events":"https://pith.science/api/pith-number/B2Y4EEKPCFSZS4PE2X6OF7XYD7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/B2Y4EEKPCFSZS4PE2X6OF7XYD7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/B2Y4EEKPCFSZS4PE2X6OF7XYD7/action/storage_attestation","attest_author":"https://pith.science/pith/B2Y4EEKPCFSZS4PE2X6OF7XYD7/action/author_attestation","sign_citation":"https://pith.science/pith/B2Y4EEKPCFSZS4PE2X6OF7XYD7/action/citation_signature","submit_replication":"https://pith.science/pith/B2Y4EEKPCFSZS4PE2X6OF7XYD7/action/replication_record"}},"created_at":"2026-07-05T09:56:22.254773+00:00","updated_at":"2026-07-05T09:56:22.254773+00:00"}