{"id":"https://openalex.org/W7166887722","doi":"https://doi.org/10.18653/v1/2026.acl-long.857","title":"Closing the Modality Reasoning Gap for Speech Large Language Models","display_name":"Closing the Modality Reasoning Gap for Speech Large Language Models","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7166887722","doi":"https://doi.org/10.18653/v1/2026.acl-long.857"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.acl-long.857","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.857","pdf_url":"https://aclanthology.org/2026.acl-long.857.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.acl-long.857.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5005342759","display_name":"Chaoren Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chaoren Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139795294","display_name":"Heng Lu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Heng Lu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139734031","display_name":"Xueyao Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xueyao Zhang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139793246","display_name":"Shujie Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shujie Liu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139793146","display_name":"Yan Lu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yan Lu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139745656","display_name":"Jinyu Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jinyu Li","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5139739699","display_name":"Zhizheng Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhizheng Wu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.87610384,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"18821","last_page":"18835"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.2736000120639801,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.2736000120639801,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12031","display_name":"Speech and dialogue systems","score":0.2685000002384186,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.12610000371932983,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/closing","display_name":"Closing (real estate)","score":0.7486000061035156},{"id":"https://openalex.org/keywords/modality","display_name":"Modality (human\u2013computer interaction)","score":0.4977000057697296},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.3824999928474426},{"id":"https://openalex.org/keywords/natural-language","display_name":"Natural language","score":0.34220001101493835},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.3287999927997589},{"id":"https://openalex.org/keywords/language-understanding","display_name":"Language understanding","score":0.2858000099658966}],"concepts":[{"id":"https://openalex.org/C2778775528","wikidata":"https://www.wikidata.org/wiki/Q5135432","display_name":"Closing (real estate)","level":2,"score":0.7486000061035156},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.641700029373169},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5254999995231628},{"id":"https://openalex.org/C2780226545","wikidata":"https://www.wikidata.org/wiki/Q6888030","display_name":"Modality (human\u2013computer interaction)","level":2,"score":0.4977000057697296},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4839000105857849},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.3824999928474426},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.351500004529953},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.34220001101493835},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.3287999927997589},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.32350000739097595},{"id":"https://openalex.org/C2983448237","wikidata":"https://www.wikidata.org/wiki/Q1078276","display_name":"Language understanding","level":2,"score":0.2858000099658966},{"id":"https://openalex.org/C2778751112","wikidata":"https://www.wikidata.org/wiki/Q835016","display_name":"Window (computing)","level":2,"score":0.27379998564720154},{"id":"https://openalex.org/C195344581","wikidata":"https://www.wikidata.org/wiki/Q2555318","display_name":"Automated reasoning","level":2,"score":0.2728999853134155},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.2587999999523163},{"id":"https://openalex.org/C155092808","wikidata":"https://www.wikidata.org/wiki/Q182557","display_name":"Computational linguistics","level":2,"score":0.2547999918460846},{"id":"https://openalex.org/C48677424","wikidata":"https://www.wikidata.org/wiki/Q6888088","display_name":"Mode (computer interface)","level":2,"score":0.2538999915122986},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.25360000133514404}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.acl-long.857","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.857","pdf_url":"https://aclanthology.org/2026.acl-long.857.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.acl-long.857","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.857","pdf_url":"https://aclanthology.org/2026.acl-long.857.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education","score":0.5101436972618103}],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166887722.pdf","grobid_xml":"https://content.openalex.org/works/W7166887722.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Although":[0],"Speech":[1,119],"Large":[2],"Language":[3],"Models":[4],"have":[5],"achieved":[6],"notable":[7],"progress,":[8],"a":[9,49],"substantial":[10],"modality":[11,110],"reasoning":[12,16,97,111],"gap":[13,27,112],"remains:":[14],"their":[15],"performance":[17,116],"on":[18,25,95],"speech":[19],"inputs":[20],"is":[21],"markedly":[22],"weaker":[23],"than":[24],"text.This":[26],"could":[28],"be":[29],"associated":[30],"with":[31],"representational":[32],"drift":[33],"across":[34],"Transformer":[35],"layers":[36],"and":[37,55,67,81,91,101,113],"behavior":[38,82],"deviations":[39],"in":[40],"long-chain":[41],"reasoning.To":[42],"address":[43],"this":[44],"issue,":[45],"we":[46],"introduce":[47],"TARS,":[48],"reinforcement-learning":[50],"framework":[51,63],"that":[52,104],"aligns":[53],"text-conditioned":[54,79],"speech-conditioned":[56],"trajectories":[57],"through":[58],"an":[59],"asymmetric":[60],"reward":[61],"design.The":[62],"employs":[64],"two":[65],"dense":[66],"complementary":[68],"signals:":[69],"representation":[70],"alignment,":[71,83],"which":[72,84],"measures":[73],"layer-wise":[74],"hidden-state":[75],"similarity":[76],"between":[77,88],"speech-and":[78],"trajectories,":[80],"evaluates":[85],"semantic":[86],"consistency":[87],"generated":[89],"outputs":[90],"reference":[92],"text":[93],"completions.Experiments":[94],"challenging":[96],"benchmarks,":[98],"including":[99],"MMSU":[100],"OBQA,":[102],"show":[103],"our":[105],"approach":[106],"significantly":[107],"narrows":[108],"the":[109],"achieves":[114],"state-of-the-art":[115],"among":[117],"7B-scale":[118],"LLMs.":[120]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-07-02T00:00:00"}