{"id":"https://openalex.org/W7166800304","doi":"https://doi.org/10.18653/v1/2026.findings-acl.2162","title":"Beyond Unimodal Shortcuts: MLLMs as Cross-Modal Reasoners for Grounded Named Entity Recognition","display_name":"Beyond Unimodal Shortcuts: MLLMs as Cross-Modal Reasoners for Grounded Named Entity Recognition","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7166800304","doi":"https://doi.org/10.18653/v1/2026.findings-acl.2162"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.findings-acl.2162","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.2162","pdf_url":"https://aclanthology.org/2026.findings-acl.2162.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":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.findings-acl.2162.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5139751520","display_name":"Jinlong Ma","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jinlong Ma","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139749190","display_name":"Yu Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yu Zhang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139819074","display_name":"Xuefeng Bai","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xuefeng Bai","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139726055","display_name":"Kehai Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kehai Chen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139812197","display_name":"Yuwei Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yuwei Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139828160","display_name":"Zeming Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zeming Liu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139749602","display_name":"Jun Yu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jun Yu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5139762730","display_name":"Min Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Min Zhang","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":0.78968507,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"43518","last_page":"43539"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.8921999931335449,"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/T10028","display_name":"Topic Modeling","score":0.8921999931335449,"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":0.019500000402331352,"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":0.012299999594688416,"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/feature","display_name":"Feature (linguistics)","score":0.3222000002861023},{"id":"https://openalex.org/keywords/named-entity-recognition","display_name":"Named-entity recognition","score":0.3012000024318695},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.26649999618530273},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.2502000033855438},{"id":"https://openalex.org/keywords/sequence","display_name":"Sequence (biology)","score":0.24199999868869781}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.589900016784668},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5142999887466431},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4555000066757202},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.3222000002861023},{"id":"https://openalex.org/C2779135771","wikidata":"https://www.wikidata.org/wiki/Q403574","display_name":"Named-entity recognition","level":3,"score":0.3012000024318695},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.26649999618530273},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.2502000033855438},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.24199999868869781},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.2386000007390976},{"id":"https://openalex.org/C116834253","wikidata":"https://www.wikidata.org/wiki/Q2039217","display_name":"Identification (biology)","level":2,"score":0.2370000034570694}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.findings-acl.2162","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.2162","pdf_url":"https://aclanthology.org/2026.findings-acl.2162.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":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.findings-acl.2162","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.2162","pdf_url":"https://aclanthology.org/2026.findings-acl.2162.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":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1289754401","display_name":null,"funder_award_id":"U24A20328","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G1956858627","display_name":null,"funder_award_id":"2026A1515011718","funder_id":"https://openalex.org/F4320337111","funder_display_name":"Basic and Applied Basic Research Foundation of Guangdong Province"},{"id":"https://openalex.org/G2263305402","display_name":null,"funder_award_id":"U23B2055","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G3537643952","display_name":null,"funder_award_id":"62350710797","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G377210702","display_name":null,"funder_award_id":"62406091","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G7320519987","display_name":null,"funder_award_id":"62276077","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G7322739430","display_name":null,"funder_award_id":"2024A1515011205","funder_id":"https://openalex.org/F4320337111","funder_display_name":"Basic and Applied Basic Research Foundation of Guangdong Province"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320337111","display_name":"Basic and Applied Basic Research Foundation of Guangdong Province","ror":null}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166800304.pdf","grobid_xml":"https://content.openalex.org/works/W7166800304.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Grounded":[0],"Multimodal":[1,29],"Named":[2],"Entity":[3],"Recognition":[4],"(GMNER)":[5],"aims":[6],"to":[7,18,34,73,118,147],"extract":[8],"text-based":[9],"entities,":[10],"assign":[11],"them":[12,17],"semantic":[13],"categories,":[14],"and":[15,65,101,131,142,151],"ground":[16],"corresponding":[19],"visual":[20,63,132],"regions.In":[21],"this":[22],"work,":[23],"we":[24,84],"explore":[25],"the":[26,116],"potential":[27],"of":[28],"Large":[30],"Language":[31],"Models":[32],"(MLLMs)":[33],"perform":[35],"GMNER":[36,130],"in":[37],"an":[38],"end-to-end":[39],"manner,":[40],"moving":[41],"beyond":[42],"their":[43,71],"typical":[44],"role":[45],"as":[46],"auxiliary":[47],"tools":[48],"within":[49],"cascaded":[50],"pipelines.Crucially,":[51],"our":[52],"investigation":[53],"reveals":[54],"a":[55],"fundamental":[56],"challenge:":[57],"MLLMs":[58],"exhibit":[59],"modality":[60,140],"bias,":[61,67],"including":[62],"bias":[64,141],"textual":[66],"which":[68,90],"stems":[69],"from":[70],"tendency":[72],"take":[74],"unimodal":[75],"shortcuts":[76],"rather":[77],"than":[78],"rigorous":[79],"cross-modal":[80,93],"verification.To":[81],"address":[82],"this,":[83],"propose":[85],"Modality-aware":[86],"Consistency":[87],"Reasoning":[88,97],"(MCR),":[89],"enforces":[91],"structured":[92],"reasoning":[94,111,122],"through":[95],"Multi-style":[96],"Schema":[98],"Injection":[99],"(MRSI)":[100],"Constraint-guided":[102],"Verifiable":[103],"Optimization":[104],"(CVO).MRSI":[105],"transforms":[106],"abstract":[107],"constraints":[108],"into":[109],"executable":[110],"chains,":[112],"while":[113],"CVO":[114],"empowers":[115],"model":[117],"dynamically":[119],"align":[120],"its":[121],"trajectories":[123],"with":[124],"Group":[125],"Relative":[126],"Policy":[127],"Optimization.Experiments":[128],"on":[129],"grounding":[133],"tasks":[134],"demonstrate":[135],"that":[136],"MCR":[137],"effectively":[138],"mitigates":[139],"achieves":[143],"superior":[144],"performance":[145],"compared":[146],"existing":[148],"baselines.The":[149],"code":[150],"data":[152],"are":[153],"released":[154],"at":[155],"https://github.com/":[156],"aaaalonga/MCR.":[157]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-07-02T00:00:00"}