{"id":"https://openalex.org/W4412888030","doi":"https://doi.org/10.18653/v1/2025.findings-acl.979","title":"From Observation to Understanding: Front-Door Adjustments with Uncertainty Calibration for Enhancing Egocentric Reasoning in LVLMs","display_name":"From Observation to Understanding: Front-Door Adjustments with Uncertainty Calibration for Enhancing Egocentric Reasoning in LVLMs","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412888030","doi":"https://doi.org/10.18653/v1/2025.findings-acl.979"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.findings-acl.979","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.979","pdf_url":"https://aclanthology.org/2025.findings-acl.979.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":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.findings-acl.979.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5043867916","display_name":"Shenshen Li","orcid":"https://orcid.org/0000-0002-6340-012X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shenshen Li","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Wenxin Meng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wenxin Meng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100435848","display_name":"Lei Wang","orcid":"https://orcid.org/0000-0002-0961-0441"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lei Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5045287885","display_name":"Hao Yang","orcid":"https://orcid.org/0000-0001-6431-8956"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hao Yang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5054958308","display_name":"Chong Peng","orcid":"https://orcid.org/0000-0003-0003-5126"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chong Peng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Peng Yan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Peng Yan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5074492050","display_name":"Fumin Shen","orcid":"https://orcid.org/0000-0001-7303-3231"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fumin Shen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5108085750","display_name":"Jingkuan Song","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jingkuan Song","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103039597","display_name":"Heng Tao Shen","orcid":"https://orcid.org/0000-0001-9872-8451"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Heng Tao Shen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5027890741","display_name":"Xing Xu","orcid":"https://orcid.org/0000-0003-4564-470X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xing Xu","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":true,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"19152","last_page":"19169"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10933","display_name":"Real-Time Systems Scheduling","score":0.902899980545044,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/T10933","display_name":"Real-Time Systems Scheduling","score":0.902899980545044,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/front","display_name":"Front (military)","score":0.6941155791282654},{"id":"https://openalex.org/keywords/calibration","display_name":"Calibration","score":0.666275680065155},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6245136260986328},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4621475636959076},{"id":"https://openalex.org/keywords/human\u2013computer-interaction","display_name":"Human\u2013computer interaction","score":0.40602707862854004},{"id":"https://openalex.org/keywords/computer-vision","display_name":"Computer vision","score":0.38633784651756287},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.16718530654907227},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.11716362833976746},{"id":"https://openalex.org/keywords/mechanical-engineering","display_name":"Mechanical engineering","score":0.08130857348442078}],"concepts":[{"id":"https://openalex.org/C2777551076","wikidata":"https://www.wikidata.org/wiki/Q842332","display_name":"Front (military)","level":2,"score":0.6941155791282654},{"id":"https://openalex.org/C165838908","wikidata":"https://www.wikidata.org/wiki/Q736777","display_name":"Calibration","level":2,"score":0.666275680065155},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6245136260986328},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4621475636959076},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.40602707862854004},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.38633784651756287},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.16718530654907227},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.11716362833976746},{"id":"https://openalex.org/C78519656","wikidata":"https://www.wikidata.org/wiki/Q101333","display_name":"Mechanical engineering","level":1,"score":0.08130857348442078},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-acl.979","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.979","pdf_url":"https://aclanthology.org/2025.findings-acl.979.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"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-acl.979","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.979","pdf_url":"https://aclanthology.org/2025.findings-acl.979.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":[{"score":0.699999988079071,"display_name":"Reduced inequalities","id":"https://metadata.un.org/sdg/10"}],"awards":[{"id":"https://openalex.org/G5347999826","display_name":null,"funder_award_id":"62222203","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412888030.pdf","grobid_xml":"https://content.openalex.org/works/W4412888030.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W2772917594","https://openalex.org/W2036807459","https://openalex.org/W2058170566","https://openalex.org/W2755342338","https://openalex.org/W2166024367","https://openalex.org/W3116076068","https://openalex.org/W2229312674","https://openalex.org/W2951359407","https://openalex.org/W2079911747","https://openalex.org/W1969923398"],"abstract_inverted_index":{"Recent":[0],"progress":[1],"in":[2,83,123],"large":[3],"vision-language":[4],"models":[5],"(LVLMs)":[6],"has":[7],"shown":[8],"substantial":[9],"potential":[10],"across":[11],"a":[12,56,97,153],"broad":[13],"spectrum":[14],"of":[15,73,171],"third-person":[16,29],"tasks.However,":[17],"adapting":[18],"these":[19,52,124],"LVLMs":[20,35,74,173],"to":[21,27,47,67,116,130,149],"egocentric":[22,49,70,94],"scenarios":[23],"remains":[24],"challenging":[25],"due":[26],"their":[28,45],"training":[30],"bias.Existing":[31],"methods":[32],"that":[33,163],"adapt":[34],"for":[36],"first-person":[37],"tasks":[38],"often":[39],"overlook":[40],"critical":[41],"agent-environment":[42],"interactions,":[43],"limiting":[44],"ability":[46],"perform":[48],"reasoning.To":[50],"address":[51],"challenges,":[53],"we":[54,92,101,126],"propose":[55],"novel":[57],"zero-shot":[58],"paradigm":[59],"termed":[60],"Front-Door":[61],"Adjustments":[62],"with":[63,113],"Uncertainty":[64],"Calibration":[65],"(FRUIT)":[66],"enhance":[68],"the":[69,80,118,143,147,159,169],"reasoning":[71,95],"abilities":[72],"by":[75],"simulating":[76],"human":[77],"causal":[78,99],"reasoning.Specifically,":[79],"FRUIT":[81,165],"operates":[82],"two":[84],"stages:":[85],"observation":[86],"and":[87,105],"understanding.Unlike":[88],"conventional":[89],"prompting":[90],"techniques,":[91],"formalize":[93],"using":[96],"structural":[98],"model.Then,":[100],"ground":[102],"interaction":[103],"regions":[104],"expand":[106],"them":[107],"into":[108,142],"hierarchical":[109],"visual":[110],"cues,":[111],"augmented":[112],"corresponding":[114],"captions,":[115],"form":[117],"initial":[119],"observations.To":[120],"reduce":[121],"noise":[122],"observations,":[125],"employ":[127],"uncertainty":[128],"calibration":[129],"filter":[131],"out":[132],"unreliable":[133],"information.These":[134],"refined":[135],"observations":[136],"as":[137],"mediators":[138],"are":[139],"then":[140],"incorporated":[141],"prompt":[144],"template,":[145],"guiding":[146],"model":[148],"understand":[150],"semantics":[151],"from":[152],"firstperson":[154],"perspective.Extensive":[155],"experiments":[156],"conducted":[157],"on":[158,174],"EgoThink":[160],"benchmark":[161],"demonstrate":[162],"our":[164],"method":[166],"consistently":[167],"enhances":[168],"performance":[170],"existing":[172],"six":[175],"distinct":[176],"tasks.Our":[177],"code":[178],"is":[179],"available":[180],"at":[181],"https://github.com/Mrshenshen/FRUIT.":[182]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}