{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:OTE366YUKCRREUOCZOSLYV2YCS","short_pith_number":"pith:OTE366YU","schema_version":"1.0","canonical_sha256":"74c9bf7b1450a31251c2cba4bc575814af1aba7ef1a134fc080aae9674827517","source":{"kind":"arxiv","id":"2403.19709","version":1},"attestation_state":"computed","paper":{"title":"Hierarchical Recurrent Adapters for Efficient Multi-Task Adaptation of Large Speech Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG","cs.NE"],"primary_cat":"eess.AS","authors_text":"Fadi Biadsy, Khe Chai Sim, Pedro Moreno Mengibar, Tara Sainath, Tsendsuren Munkhdalai, Youzheng Chen","submitted_at":"2024-03-25T17:21:56Z","abstract_excerpt":"Parameter efficient adaptation methods have become a key mechanism to train large pre-trained models for downstream tasks. However, their per-task parameter overhead is considered still high when the number of downstream tasks to adapt for is large. We introduce an adapter module that has a better efficiency in large scale multi-task adaptation scenario. Our adapter is hierarchical in terms of how the adapter parameters are allocated. The adapter consists of a single shared controller network and multiple task-level adapter heads to reduce the per-task parameter overhead without performance re"},"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":"2403.19709","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2024-03-25T17:21:56Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG","cs.NE"],"title_canon_sha256":"a91b82751d1ed349df7579e6ba4f26d4dc3bd1e05c4d47a0da194371a89b4e60","abstract_canon_sha256":"45cfb978222e0834951c11033684bdde9d9c6be21853cb6085c76c3de366aacd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:02:07.464135Z","signature_b64":"SXnQVPmkXTwDWoSXyJEaQhcsxz3XeibUridorPI8LlerUxi+OaueVCxJC7bPUqGeXl3TrMmK/Pe0XaqKq1bMDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"74c9bf7b1450a31251c2cba4bc575814af1aba7ef1a134fc080aae9674827517","last_reissued_at":"2026-07-05T08:02:07.463736Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:02:07.463736Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Hierarchical Recurrent Adapters for Efficient Multi-Task Adaptation of Large Speech Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG","cs.NE"],"primary_cat":"eess.AS","authors_text":"Fadi Biadsy, Khe Chai Sim, Pedro Moreno Mengibar, Tara Sainath, Tsendsuren Munkhdalai, Youzheng Chen","submitted_at":"2024-03-25T17:21:56Z","abstract_excerpt":"Parameter efficient adaptation methods have become a key mechanism to train large pre-trained models for downstream tasks. However, their per-task parameter overhead is considered still high when the number of downstream tasks to adapt for is large. We introduce an adapter module that has a better efficiency in large scale multi-task adaptation scenario. Our adapter is hierarchical in terms of how the adapter parameters are allocated. The adapter consists of a single shared controller network and multiple task-level adapter heads to reduce the per-task parameter overhead without performance re"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.19709","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/2403.19709/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":"2403.19709","created_at":"2026-07-05T08:02:07.463800+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.19709v1","created_at":"2026-07-05T08:02:07.463800+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.19709","created_at":"2026-07-05T08:02:07.463800+00:00"},{"alias_kind":"pith_short_12","alias_value":"OTE366YUKCRR","created_at":"2026-07-05T08:02:07.463800+00:00"},{"alias_kind":"pith_short_16","alias_value":"OTE366YUKCRREUOC","created_at":"2026-07-05T08:02:07.463800+00:00"},{"alias_kind":"pith_short_8","alias_value":"OTE366YU","created_at":"2026-07-05T08:02:07.463800+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/OTE366YUKCRREUOCZOSLYV2YCS","json":"https://pith.science/pith/OTE366YUKCRREUOCZOSLYV2YCS.json","graph_json":"https://pith.science/api/pith-number/OTE366YUKCRREUOCZOSLYV2YCS/graph.json","events_json":"https://pith.science/api/pith-number/OTE366YUKCRREUOCZOSLYV2YCS/events.json","paper":"https://pith.science/paper/OTE366YU"},"agent_actions":{"view_html":"https://pith.science/pith/OTE366YUKCRREUOCZOSLYV2YCS","download_json":"https://pith.science/pith/OTE366YUKCRREUOCZOSLYV2YCS.json","view_paper":"https://pith.science/paper/OTE366YU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.19709&json=true","fetch_graph":"https://pith.science/api/pith-number/OTE366YUKCRREUOCZOSLYV2YCS/graph.json","fetch_events":"https://pith.science/api/pith-number/OTE366YUKCRREUOCZOSLYV2YCS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OTE366YUKCRREUOCZOSLYV2YCS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OTE366YUKCRREUOCZOSLYV2YCS/action/storage_attestation","attest_author":"https://pith.science/pith/OTE366YUKCRREUOCZOSLYV2YCS/action/author_attestation","sign_citation":"https://pith.science/pith/OTE366YUKCRREUOCZOSLYV2YCS/action/citation_signature","submit_replication":"https://pith.science/pith/OTE366YUKCRREUOCZOSLYV2YCS/action/replication_record"}},"created_at":"2026-07-05T08:02:07.463800+00:00","updated_at":"2026-07-05T08:02:07.463800+00:00"}