{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:FYKEWG6JQ3YRILG6G2Q4LHHD2G","short_pith_number":"pith:FYKEWG6J","schema_version":"1.0","canonical_sha256":"2e144b1bc986f1142cde36a1c59ce3d1ac170517e95efa6e9178b67f9374c353","source":{"kind":"arxiv","id":"2406.05839","version":2},"attestation_state":"computed","paper":{"title":"MaLa-ASR: Multimedia-Assisted LLM-Based ASR","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"eess.AS","authors_text":"Fan Yu, Guanrou Yang, Shiliang Zhang, Xie Chen, Zhifu Gao, Ziyang Ma","submitted_at":"2024-06-09T16:00:00Z","abstract_excerpt":"As more and more information-rich data like video become available, utilizing multi-modal auxiliary information to enhance audio tasks has sparked widespread research interest. The recent surge in research on LLM-based audio models provides fresh perspectives for tackling audio tasks. Given that LLM can flexibly ingest multiple inputs, we propose MaLa-ASR, an LLM-based ASR model that can integrate textual keywords extracted from presentation slides to improve recognition of conference content. MaLa-ASR yields average WERs of 9.4% and 11.7% on the L95 and S95 subsets of the SlideSpeech corpus, "},"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":"2406.05839","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.AS","submitted_at":"2024-06-09T16:00:00Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9fdaec674e7de5ab44ab303270943260aaddf6d5d2db64b2538d818a85ddcfae","abstract_canon_sha256":"597962f8e73c40119d187228b2fe8a4a0aea7ee1bbec0d46f6282311f3157c18"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:33:32.884038Z","signature_b64":"T23RMEZEFVNI1XdB/C0SVdla2gRIKAAT8DwM2/Q/nUwpzqYv7HsdvJc90r+q3rtkr3qbFdgVFpvi38OVSW0UCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2e144b1bc986f1142cde36a1c59ce3d1ac170517e95efa6e9178b67f9374c353","last_reissued_at":"2026-07-05T09:33:32.883475Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:33:32.883475Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MaLa-ASR: Multimedia-Assisted LLM-Based ASR","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"eess.AS","authors_text":"Fan Yu, Guanrou Yang, Shiliang Zhang, Xie Chen, Zhifu Gao, Ziyang Ma","submitted_at":"2024-06-09T16:00:00Z","abstract_excerpt":"As more and more information-rich data like video become available, utilizing multi-modal auxiliary information to enhance audio tasks has sparked widespread research interest. The recent surge in research on LLM-based audio models provides fresh perspectives for tackling audio tasks. Given that LLM can flexibly ingest multiple inputs, we propose MaLa-ASR, an LLM-based ASR model that can integrate textual keywords extracted from presentation slides to improve recognition of conference content. MaLa-ASR yields average WERs of 9.4% and 11.7% on the L95 and S95 subsets of the SlideSpeech corpus, "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.05839","kind":"arxiv","version":2},"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/2406.05839/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":"2406.05839","created_at":"2026-07-05T09:33:32.883565+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.05839v2","created_at":"2026-07-05T09:33:32.883565+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.05839","created_at":"2026-07-05T09:33:32.883565+00:00"},{"alias_kind":"pith_short_12","alias_value":"FYKEWG6JQ3YR","created_at":"2026-07-05T09:33:32.883565+00:00"},{"alias_kind":"pith_short_16","alias_value":"FYKEWG6JQ3YRILG6","created_at":"2026-07-05T09:33:32.883565+00:00"},{"alias_kind":"pith_short_8","alias_value":"FYKEWG6J","created_at":"2026-07-05T09:33:32.883565+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.10456","citing_title":"Exploring Cross-Utterance Speech Contexts for Conformer-Transducer Speech Recognition Systems","ref_index":47,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FYKEWG6JQ3YRILG6G2Q4LHHD2G","json":"https://pith.science/pith/FYKEWG6JQ3YRILG6G2Q4LHHD2G.json","graph_json":"https://pith.science/api/pith-number/FYKEWG6JQ3YRILG6G2Q4LHHD2G/graph.json","events_json":"https://pith.science/api/pith-number/FYKEWG6JQ3YRILG6G2Q4LHHD2G/events.json","paper":"https://pith.science/paper/FYKEWG6J"},"agent_actions":{"view_html":"https://pith.science/pith/FYKEWG6JQ3YRILG6G2Q4LHHD2G","download_json":"https://pith.science/pith/FYKEWG6JQ3YRILG6G2Q4LHHD2G.json","view_paper":"https://pith.science/paper/FYKEWG6J","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.05839&json=true","fetch_graph":"https://pith.science/api/pith-number/FYKEWG6JQ3YRILG6G2Q4LHHD2G/graph.json","fetch_events":"https://pith.science/api/pith-number/FYKEWG6JQ3YRILG6G2Q4LHHD2G/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FYKEWG6JQ3YRILG6G2Q4LHHD2G/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FYKEWG6JQ3YRILG6G2Q4LHHD2G/action/storage_attestation","attest_author":"https://pith.science/pith/FYKEWG6JQ3YRILG6G2Q4LHHD2G/action/author_attestation","sign_citation":"https://pith.science/pith/FYKEWG6JQ3YRILG6G2Q4LHHD2G/action/citation_signature","submit_replication":"https://pith.science/pith/FYKEWG6JQ3YRILG6G2Q4LHHD2G/action/replication_record"}},"created_at":"2026-07-05T09:33:32.883565+00:00","updated_at":"2026-07-05T09:33:32.883565+00:00"}