{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:OXKBG7BC7A2SLXYQ243JM7C4SU","short_pith_number":"pith:OXKBG7BC","schema_version":"1.0","canonical_sha256":"75d4137c22f83525df10d736967c5c95353142f45b20a0281672ce019dfd1524","source":{"kind":"arxiv","id":"2408.16423","version":1},"attestation_state":"computed","paper":{"title":"WHISMA: A Speech-LLM to Perform Zero-shot Spoken Language Understanding","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.SD"],"primary_cat":"eess.AS","authors_text":"Cong-Thanh Do, Mohan Li, Rama Doddipatla, Simon Keizer, Svetlana Stoyanchev, Youmna Farag","submitted_at":"2024-08-29T10:31:52Z","abstract_excerpt":"Speech large language models (speech-LLMs) integrate speech and text-based foundation models to provide a unified framework for handling a wide range of downstream tasks. In this paper, we introduce WHISMA, a speech-LLM tailored for spoken language understanding (SLU) that demonstrates robust performance in various zero-shot settings. WHISMA combines the speech encoder from Whisper with the Llama-3 LLM, and is fine-tuned in a parameter-efficient manner on a comprehensive collection of SLU-related datasets. Our experiments show that WHISMA significantly improves the zero-shot slot filling perfo"},"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":"2408.16423","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"eess.AS","submitted_at":"2024-08-29T10:31:52Z","cross_cats_sorted":["cs.SD"],"title_canon_sha256":"c49c52df6df3d24480952b2aec0dbd5da30153c0b98ce0c7882fd229d8c0182d","abstract_canon_sha256":"234a6422b9f8c217996b83197f1d518a02b61ce456efa1999f5912a90d0420c8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:00:41.152783Z","signature_b64":"CHwzh/Mf/Pzwz+vnJrud89LTmGwFsuxtzXM0Ryrx9TwP9HvYDtlPJ98V/0bBeT5DfF6syqao3HwRnKohEOinCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"75d4137c22f83525df10d736967c5c95353142f45b20a0281672ce019dfd1524","last_reissued_at":"2026-07-05T09:00:41.152345Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:00:41.152345Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"WHISMA: A Speech-LLM to Perform Zero-shot Spoken Language Understanding","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.SD"],"primary_cat":"eess.AS","authors_text":"Cong-Thanh Do, Mohan Li, Rama Doddipatla, Simon Keizer, Svetlana Stoyanchev, Youmna Farag","submitted_at":"2024-08-29T10:31:52Z","abstract_excerpt":"Speech large language models (speech-LLMs) integrate speech and text-based foundation models to provide a unified framework for handling a wide range of downstream tasks. In this paper, we introduce WHISMA, a speech-LLM tailored for spoken language understanding (SLU) that demonstrates robust performance in various zero-shot settings. WHISMA combines the speech encoder from Whisper with the Llama-3 LLM, and is fine-tuned in a parameter-efficient manner on a comprehensive collection of SLU-related datasets. Our experiments show that WHISMA significantly improves the zero-shot slot filling perfo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.16423","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/2408.16423/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":"2408.16423","created_at":"2026-07-05T09:00:41.152401+00:00"},{"alias_kind":"arxiv_version","alias_value":"2408.16423v1","created_at":"2026-07-05T09:00:41.152401+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.16423","created_at":"2026-07-05T09:00:41.152401+00:00"},{"alias_kind":"pith_short_12","alias_value":"OXKBG7BC7A2S","created_at":"2026-07-05T09:00:41.152401+00:00"},{"alias_kind":"pith_short_16","alias_value":"OXKBG7BC7A2SLXYQ","created_at":"2026-07-05T09:00:41.152401+00:00"},{"alias_kind":"pith_short_8","alias_value":"OXKBG7BC","created_at":"2026-07-05T09:00:41.152401+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.17202","citing_title":"Audio Large Language Models Can Be Descriptive Speech Quality Evaluators","ref_index":18,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OXKBG7BC7A2SLXYQ243JM7C4SU","json":"https://pith.science/pith/OXKBG7BC7A2SLXYQ243JM7C4SU.json","graph_json":"https://pith.science/api/pith-number/OXKBG7BC7A2SLXYQ243JM7C4SU/graph.json","events_json":"https://pith.science/api/pith-number/OXKBG7BC7A2SLXYQ243JM7C4SU/events.json","paper":"https://pith.science/paper/OXKBG7BC"},"agent_actions":{"view_html":"https://pith.science/pith/OXKBG7BC7A2SLXYQ243JM7C4SU","download_json":"https://pith.science/pith/OXKBG7BC7A2SLXYQ243JM7C4SU.json","view_paper":"https://pith.science/paper/OXKBG7BC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2408.16423&json=true","fetch_graph":"https://pith.science/api/pith-number/OXKBG7BC7A2SLXYQ243JM7C4SU/graph.json","fetch_events":"https://pith.science/api/pith-number/OXKBG7BC7A2SLXYQ243JM7C4SU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OXKBG7BC7A2SLXYQ243JM7C4SU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OXKBG7BC7A2SLXYQ243JM7C4SU/action/storage_attestation","attest_author":"https://pith.science/pith/OXKBG7BC7A2SLXYQ243JM7C4SU/action/author_attestation","sign_citation":"https://pith.science/pith/OXKBG7BC7A2SLXYQ243JM7C4SU/action/citation_signature","submit_replication":"https://pith.science/pith/OXKBG7BC7A2SLXYQ243JM7C4SU/action/replication_record"}},"created_at":"2026-07-05T09:00:41.152401+00:00","updated_at":"2026-07-05T09:00:41.152401+00:00"}