{"id":"https://openalex.org/W7166874876","doi":"https://doi.org/10.18653/v1/2026.acl-long.1699","title":"The Dominance of Text Space: Unveiling the Asymmetric Nature of Cross-Modal Alignment in Large Language Models","display_name":"The Dominance of Text Space: Unveiling the Asymmetric Nature of Cross-Modal Alignment in Large Language Models","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7166874876","doi":"https://doi.org/10.18653/v1/2026.acl-long.1699"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.acl-long.1699","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.1699","pdf_url":"https://aclanthology.org/2026.acl-long.1699.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.1699.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5011874659","display_name":"Linqing Chen","orcid":"https://orcid.org/0000-0002-8178-6636"},"institutions":[{"id":"https://openalex.org/I2534966","display_name":"Patna University","ror":"https://ror.org/04ysp6769","country_code":"IN","type":"education","lineage":["https://openalex.org/I2534966"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Linqing Chen","raw_affiliation_strings":["PatSnap Co., LTD. Suzhou , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"PatSnap Co., LTD. Suzhou , China","institution_ids":["https://openalex.org/I2534966"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139827007","display_name":"Hanmeng Zhong","orcid":null},"institutions":[{"id":"https://openalex.org/I2534966","display_name":"Patna University","ror":"https://ror.org/04ysp6769","country_code":"IN","type":"education","lineage":["https://openalex.org/I2534966"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Hanmeng Zhong","raw_affiliation_strings":["PatSnap Co., LTD. Suzhou , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"PatSnap Co., LTD. Suzhou , China","institution_ids":["https://openalex.org/I2534966"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139826085","display_name":"Wentao Wu","orcid":null},"institutions":[{"id":"https://openalex.org/I2534966","display_name":"Patna University","ror":"https://ror.org/04ysp6769","country_code":"IN","type":"education","lineage":["https://openalex.org/I2534966"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Wentao Wu","raw_affiliation_strings":["PatSnap Co., LTD. Suzhou , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"PatSnap Co., LTD. Suzhou , China","institution_ids":["https://openalex.org/I2534966"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5139801471","display_name":"Peng Zhou","orcid":null},"institutions":[{"id":"https://openalex.org/I2534966","display_name":"Patna University","ror":"https://ror.org/04ysp6769","country_code":"IN","type":"education","lineage":["https://openalex.org/I2534966"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Peng Zhou","raw_affiliation_strings":["PatSnap Co., LTD. Suzhou , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"PatSnap Co., LTD. Suzhou , China","institution_ids":["https://openalex.org/I2534966"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I2534966"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.9137931,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"36666","last_page":"36678"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12090","display_name":"Language and cultural evolution","score":0.47440001368522644,"subfield":{"id":"https://openalex.org/subfields/3316","display_name":"Cultural Studies"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T12090","display_name":"Language and cultural evolution","score":0.47440001368522644,"subfield":{"id":"https://openalex.org/subfields/3316","display_name":"Cultural Studies"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T12380","display_name":"Authorship Attribution and Profiling","score":0.2167000025510788,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.05389999970793724,"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/dominance","display_name":"Dominance (genetics)","score":0.36899998784065247},{"id":"https://openalex.org/keywords/natural-language","display_name":"Natural language","score":0.3580999970436096},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.32170000672340393},{"id":"https://openalex.org/keywords/human-language","display_name":"Human language","score":0.2791999876499176}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5188999772071838},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.46650001406669617},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4652000069618225},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.43880000710487366},{"id":"https://openalex.org/C151913843","wikidata":"https://www.wikidata.org/wiki/Q3454555","display_name":"Dominance (genetics)","level":3,"score":0.36899998784065247},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.3580999970436096},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.32170000672340393},{"id":"https://openalex.org/C46312422","wikidata":"https://www.wikidata.org/wiki/Q11024","display_name":"Communication","level":1,"score":0.2809000015258789},{"id":"https://openalex.org/C2993724205","wikidata":"https://www.wikidata.org/wiki/Q315","display_name":"Human language","level":2,"score":0.2791999876499176},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.24860000610351562}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.acl-long.1699","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.1699","pdf_url":"https://aclanthology.org/2026.acl-long.1699.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.1699","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.1699","pdf_url":"https://aclanthology.org/2026.acl-long.1699.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":[{"score":0.61687171459198,"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166874876.pdf","grobid_xml":"https://content.openalex.org/works/W7166874876.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Recent":[0],"advancements":[1],"in":[2,99],"Multimodal":[3],"LargeLanguage":[4],"Models":[5,19],"(MLLMs)":[6],"have":[7],"largely":[8],"been":[9],"driven":[10],"by":[11],"aligning":[12],"visual":[13,37,85,106],"encoders":[14],"with":[15],"pre-trained":[16],"Large":[17],"Language":[18],"(LLMs).While":[20],"effective,":[21],"the":[22,57,65,110,118,134,138,142,161],"geometric":[23,139],"nature":[24],"of":[25,68,84,117,141],"this":[26,49,53,88],"alignment":[27,78,119],"remains":[28],"under-explored.Existing":[29],"methods":[30],"often":[31],"assume":[32],"a":[33,96,129],"symmetric":[34],"interaction":[35],"between":[36],"and":[38,55,73,153,169],"textual":[39],"modalities,":[40],"implying":[41],"that":[42,64,132,158],"both":[43],"spaces":[44],"adapt":[45],"to":[46,114,136],"each":[47],"other.In":[48],"paper,":[50],"we":[51,124],"challenge":[52],"assumption":[54],"propose":[56],"\"Text":[58],"Space":[59],"as":[60],"Anchor\"":[61],"hypothesis.We":[62],"argue":[63],"semantic":[66],"space":[67,145],"LLMs":[69],"is":[70,180],"rigid,":[71],"anisotropic,":[72],"dominant;":[74],"thus,":[75],"effective":[76],"cross-modal":[77],"may":[79],"be":[80],"an":[81],"asymmetric":[82],"projection":[83],"features":[86],"onto":[87],"preexisting":[89],"text":[90,143],"manifold":[91],"without":[92],"distorting":[93],"it.We":[94],"identify":[95],"potential":[97],"issue":[98],"current":[100],"parameterefficient":[101],"tuning":[102],"paradigms":[103],"where":[104],"task-specific":[105],"adjustments":[107],"inadvertently":[108],"disrupt":[109],"projector's":[111],"geometry,":[112],"leading":[113],"\"catastrophic":[115],"forgetting\"":[116],"mechanism":[120],"itself.To":[121],"address":[122],"this,":[123],"introduce":[125],"Anchor-Preserving":[126],"Projection":[127],"(APP),":[128],"novel":[130],"method":[131],"regularizes":[133],"projector":[135],"maintain":[137],"structure":[140],"embedding":[144],"via":[146],"spectral":[147],"filtering.Extensive":[148],"experiments":[149],"on":[150],"crossmodal":[151],"tasks":[152],"pure":[154],"language":[155],"benchmarks":[156],"demonstrate":[157],"APP":[159],"preserves":[160],"LLM's":[162],"inherent":[163],"linguistic":[164],"capabilities":[165],"(e.g.,":[166],"MMLU,":[167],"GSM8K)":[168],"reduces":[170],"object":[171],"hallucination":[172],"significantly":[173],"better":[174],"than":[175],"standard":[176],"fine-tuning":[177],"methods.Our":[178],"code":[179],"publicly":[181],"available":[182],"1":[183],".":[184]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-07-02T00:00:00"}