{"id":"https://openalex.org/W4412888178","doi":"https://doi.org/10.18653/v1/2025.findings-acl.837","title":"Training Multi-Modal LLMs through Dialogue Planning for HRI","display_name":"Training Multi-Modal LLMs through Dialogue Planning for HRI","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412888178","doi":"https://doi.org/10.18653/v1/2025.findings-acl.837"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.findings-acl.837","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.837","pdf_url":"https://aclanthology.org/2025.findings-acl.837.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 2025","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.findings-acl.837.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5002939363","display_name":"Claudiu Daniel Hromei","orcid":"https://orcid.org/0009-0000-8204-5023"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Claudiu Daniel Hromei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5095839162","display_name":"Federico Borazio","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Federico Borazio","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5119181351","display_name":"Andrea Sensi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Andrea Sensi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5115717187","display_name":"Elisa Passone","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Elisa Passone","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5041324718","display_name":"Danilo Croce","orcid":"https://orcid.org/0000-0001-9111-1950"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Danilo Croce","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5083592532","display_name":"Roberto Basili","orcid":"https://orcid.org/0000-0001-5140-0694"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Roberto Basili","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":2.792,"has_fulltext":true,"cited_by_count":2,"citation_normalized_percentile":{"value":0.91220716,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":91,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"16266","last_page":"16284"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12031","display_name":"Speech and dialogue systems","score":0.9700000286102295,"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/T12031","display_name":"Speech and dialogue systems","score":0.9700000286102295,"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/modal","display_name":"Modal","score":0.6187124848365784},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.554624080657959},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.48519375920295715},{"id":"https://openalex.org/keywords/geography","display_name":"Geography","score":0.10102993249893188},{"id":"https://openalex.org/keywords/meteorology","display_name":"Meteorology","score":0.07168051600456238},{"id":"https://openalex.org/keywords/materials-science","display_name":"Materials science","score":0.06373733282089233}],"concepts":[{"id":"https://openalex.org/C71139939","wikidata":"https://www.wikidata.org/wiki/Q910194","display_name":"Modal","level":2,"score":0.6187124848365784},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.554624080657959},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.48519375920295715},{"id":"https://openalex.org/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.10102993249893188},{"id":"https://openalex.org/C153294291","wikidata":"https://www.wikidata.org/wiki/Q25261","display_name":"Meteorology","level":1,"score":0.07168051600456238},{"id":"https://openalex.org/C192562407","wikidata":"https://www.wikidata.org/wiki/Q228736","display_name":"Materials science","level":0,"score":0.06373733282089233},{"id":"https://openalex.org/C188027245","wikidata":"https://www.wikidata.org/wiki/Q750446","display_name":"Polymer chemistry","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.18653/v1/2025.findings-acl.837","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.837","pdf_url":"https://aclanthology.org/2025.findings-acl.837.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 2025","raw_type":"proceedings-article"},{"id":"pmh:oai:art.torvergata.it:2108/455226","is_oa":false,"landing_page_url":"https://hdl.handle.net/2108/455226","pdf_url":null,"source":{"id":"https://openalex.org/S4306400993","display_name":"Cineca Institutional Research Information System (Tor Vergata University)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I116067653","host_organization_name":"University of Rome Tor Vergata","host_organization_lineage":["https://openalex.org/I116067653"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"info:eu-repo/semantics/conferenceObject"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-acl.837","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.837","pdf_url":"https://aclanthology.org/2025.findings-acl.837.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 2025","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G4508289328","display_name":null,"funder_award_id":"PE00000013","funder_id":"https://openalex.org/F4320320300","funder_display_name":"European Commission"},{"id":"https://openalex.org/G4971014005","display_name":null,"funder_award_id":"CUP B83C22002820006","funder_id":"https://openalex.org/F4320320300","funder_display_name":"European Commission"},{"id":"https://openalex.org/G5019181742","display_name":null,"funder_award_id":"ECS 0000024","funder_id":"https://openalex.org/F4320320300","funder_display_name":"European Commission"},{"id":"https://openalex.org/G6025850787","display_name":null,"funder_award_id":"B83C22002820006","funder_id":"https://openalex.org/F4320320300","funder_display_name":"European Commission"},{"id":"https://openalex.org/G8363454223","display_name":null,"funder_award_id":"0000024","funder_id":"https://openalex.org/F4320320300","funder_display_name":"European Commission"}],"funders":[{"id":"https://openalex.org/F4320320300","display_name":"European Commission","ror":"https://ror.org/00k4n6c32"},{"id":"https://openalex.org/F4320322144","display_name":"Universit\u00e0 Campus Bio-Medico di Roma","ror":"https://ror.org/04gqx4x78"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412888178.pdf","grobid_xml":"https://content.openalex.org/works/W4412888178.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W230091440","https://openalex.org/W2390279801","https://openalex.org/W2233261550","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2810751659"],"abstract_inverted_index":{"Grounded":[0],"natural":[1],"language":[2,106],"understanding":[3,44],"in":[4,72,83,104,129],"Human-Robot":[5],"Interaction":[6],"(HRI)":[7],"requires":[8],"integrating":[9],"linguistic,":[10],"visual,":[11],"and":[12,54,74,79,94,96,125],"world":[13],"knowledge":[14],"to":[15,40,98,107,110],"ensure":[16],"effective":[17],"task":[18,56],"execution.We":[19],"propose":[20],"an":[21],"approach":[22],"that":[23,88],"enhances":[24],"Multi-Modal":[25,81,130],"Large":[26],"Language":[27],"Models":[28],"(MLLMs)":[29],"with":[30],"a":[31,63],"novel":[32,64],"explicit":[33,126],"dialogue":[34,89,127],"planning":[35,90,128],"phase,":[36],"allowing":[37],"robotic":[38],"agents":[39],"systematically":[41],"refine":[42],"their":[43],"of":[45,66,114,122],"ambiguous":[46],"commands":[47],"through":[48],"structured":[49],"clarification":[50],"steps.This":[51],"reduces":[52],"hallucinations":[53],"improves":[55,91],"feasibility.To":[57],"evaluate":[58],"this":[59,117],"approach,":[60],"we":[61],"introduce":[62],"dataset":[65],"over":[67],"1,":[68],"100":[69],"annotated":[70],"dialogues":[71],"English":[73],"Italian,":[75],"designed":[76],"for":[77,132],"fine-tuning":[78],"assessing":[80],"models":[82,102],"HRI":[84],"scenarios.Experimental":[85],"results":[86],"show":[87],"response":[92],"accuracy":[93],"quality,":[95],"contributes":[97],"cross-lingual":[99],"generalisation,":[100],"enabling":[101],"trained":[103],"one":[105],"transfer":[108],"effectively":[109],"another.To":[111],"the":[112,119],"best":[113],"our":[115],"knowledge,":[116],"is":[118],"first":[120],"application":[121],"structured,":[123],"goal-driven,":[124],"LLMs":[131],"grounded":[133],"interaction.":[134]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}