{"id":"https://openalex.org/W7127300873","doi":"https://doi.org/10.48550/arxiv.2602.02063","title":"See2Refine: Vision-Language Feedback Improves LLM-Based eHMI Action Designers","display_name":"See2Refine: Vision-Language Feedback Improves LLM-Based eHMI Action Designers","publication_year":2026,"publication_date":"2026-02-02","ids":{"openalex":"https://openalex.org/W7127300873","doi":"https://doi.org/10.48550/arxiv.2602.02063"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2602.02063","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.02063","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2602.02063","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5124970547","display_name":"Ding Xia","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xia, Ding","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5062839476","display_name":"Xinyue Gui","orcid":"https://orcid.org/0000-0001-6541-224X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gui, Xinyue","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5026729075","display_name":"Mark Colley","orcid":"https://orcid.org/0000-0001-5207-5029"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Colley, Mark","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5124891244","display_name":"Fan Gao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gao, Fan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5022010343","display_name":"Zhongyi Zhou","orcid":"https://orcid.org/0000-0002-1363-7313"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Zhongyi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5047165324","display_name":"Dongyuan Li","orcid":"https://orcid.org/0000-0002-4462-3563"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Dongyuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5123964335","display_name":"Renhe Jiang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiang, Renhe","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5102743150","display_name":"Takeo Igarashi","orcid":"https://orcid.org/0000-0002-5495-6441"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Igarashi, Takeo","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":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10525","display_name":"Human-Automation Interaction and Safety","score":0.762499988079071,"subfield":{"id":"https://openalex.org/subfields/3207","display_name":"Social Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T10525","display_name":"Human-Automation Interaction and Safety","score":0.762499988079071,"subfield":{"id":"https://openalex.org/subfields/3207","display_name":"Social Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T10883","display_name":"Ethics and Social Impacts of AI","score":0.07129999995231628,"subfield":{"id":"https://openalex.org/subfields/3311","display_name":"Safety Research"},"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/T11099","display_name":"Autonomous Vehicle Technology and Safety","score":0.024700000882148743,"subfield":{"id":"https://openalex.org/subfields/2203","display_name":"Automotive Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/modalities","display_name":"Modalities","score":0.7045999765396118},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.6541000008583069},{"id":"https://openalex.org/keywords/perception","display_name":"Perception","score":0.5332000255584717},{"id":"https://openalex.org/keywords/action","display_name":"Action (physics)","score":0.517799973487854},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.47269999980926514},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.4569999873638153},{"id":"https://openalex.org/keywords/natural-language","display_name":"Natural language","score":0.3278999924659729}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.781000018119812},{"id":"https://openalex.org/C2779903281","wikidata":"https://www.wikidata.org/wiki/Q6888026","display_name":"Modalities","level":2,"score":0.7045999765396118},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.6541000008583069},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.6399000287055969},{"id":"https://openalex.org/C26760741","wikidata":"https://www.wikidata.org/wiki/Q160402","display_name":"Perception","level":2,"score":0.5332000255584717},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.517799973487854},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.47269999980926514},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.4569999873638153},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.421999990940094},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.34380000829696655},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.3278999924659729},{"id":"https://openalex.org/C3020716817","wikidata":"https://www.wikidata.org/wiki/Q4132092","display_name":"Visual feedback","level":2,"score":0.3125999867916107},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.28519999980926514},{"id":"https://openalex.org/C2780226545","wikidata":"https://www.wikidata.org/wiki/Q6888030","display_name":"Modality (human\u2013computer interaction)","level":2,"score":0.28380000591278076},{"id":"https://openalex.org/C115901376","wikidata":"https://www.wikidata.org/wiki/Q184199","display_name":"Automation","level":2,"score":0.27469998598098755},{"id":"https://openalex.org/C2987834672","wikidata":"https://www.wikidata.org/wiki/Q4677630","display_name":"Action recognition","level":3,"score":0.26339998841285706},{"id":"https://openalex.org/C178253425","wikidata":"https://www.wikidata.org/wiki/Q162668","display_name":"Visual perception","level":3,"score":0.26089999079704285},{"id":"https://openalex.org/C2776608160","wikidata":"https://www.wikidata.org/wiki/Q4785462","display_name":"Natural (archaeology)","level":2,"score":0.2574000060558319}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2602.02063","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.02063","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2602.02063","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.02063","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Automated":[0],"vehicles":[1],"lack":[2,66],"natural":[3],"communication":[4],"channels":[5],"with":[6,194],"other":[7],"road":[8],"users,":[9],"making":[10],"external":[11],"Human-Machine":[12],"Interfaces":[13],"(eHMIs)":[14],"essential":[15],"for":[16,79,204],"conveying":[17],"intent":[18],"and":[19,41,69,110,124,151,153,167,175,187,200],"maintaining":[20],"trust":[21],"in":[22,171],"shared":[23],"environments.":[24],"However,":[25],"most":[26],"eHMI":[27,61,103,113,147],"studies":[28],"rely":[29],"on":[30,72],"developer-crafted":[31],"message-action":[32],"pairs,":[33],"which":[34],"are":[35,191],"difficult":[36],"to":[37,39,49,99,129],"adapt":[38],"diverse":[40],"dynamic":[42],"traffic":[43],"contexts.":[44],"A":[45],"promising":[46],"alternative":[47],"is":[48,127],"use":[50],"Large":[51],"Language":[52],"Models":[53],"(LLMs)":[54],"as":[55,95],"action":[56,104,206],"designers":[57,65,166],"that":[58,88,181,188],"generate":[59],"context-conditioned":[60],"actions,":[62],"yet":[63],"such":[64],"perceptual":[67,93],"verification":[68],"typically":[70],"depend":[71],"fixed":[73],"prompts":[74],"or":[75],"costly":[76],"human-annotated":[77],"feedback":[78,98,126],"improvement.":[80],"We":[81,141],"present":[82],"See2Refine,":[83],"a":[84,107,111],"human-free,":[85],"closed-loop":[86],"framework":[87,144,161],"uses":[89],"vision-language":[90],"model":[91,156],"(VLM)":[92],"evaluation":[94],"automated":[96],"visual":[97],"improve":[100],"an":[101],"LLM-based":[102],"designer.":[105],"Given":[106],"driving":[108],"context":[109],"candidate":[112],"action,":[114,123],"the":[115,118,122,132,182,198],"VLM":[116,189],"evaluates":[117],"perceived":[119],"appropriateness":[120],"of":[121,202],"this":[125],"used":[128],"iteratively":[130],"revise":[131],"designer's":[133],"outputs,":[134],"enabling":[135],"systematic":[136],"refinement":[137],"without":[138],"human":[139,195],"supervision.":[140],"evaluate":[142],"our":[143,160],"across":[145,185],"three":[146],"modalities":[148,186],"(lightbar,":[149],"eyes,":[150],"arm)":[152],"multiple":[154],"LLM":[155,165],"sizes.":[157],"Across":[158],"settings,":[159],"consistently":[162],"outperforms":[163],"prompt-only":[164],"manually":[168],"specified":[169],"baselines":[170],"both":[172],"VLM-based":[173],"metrics":[174],"human-subject":[176],"evaluations.":[177],"Results":[178],"further":[179],"indicate":[180],"improvements":[183],"generalize":[184],"evaluations":[190],"well":[192],"aligned":[193],"preferences,":[196],"supporting":[197],"robustness":[199],"effectiveness":[201],"See2Refine":[203],"scalable":[205],"design.":[207]},"counts_by_year":[],"updated_date":"2026-08-16T07:02:28.622633","created_date":"2026-02-04T00:00:00"}