{"id":"https://openalex.org/W4412945065","doi":"https://doi.org/10.18653/v1/2025.acl-long.802","title":"MMBoundary: Advancing MLLM Knowledge Boundary Awareness through Reasoning Step Confidence Calibration","display_name":"MMBoundary: Advancing MLLM Knowledge Boundary Awareness through Reasoning Step Confidence Calibration","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412945065","doi":"https://doi.org/10.18653/v1/2025.acl-long.802"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.acl-long.802","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.802","pdf_url":"https://aclanthology.org/2025.acl-long.802.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 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.acl-long.802.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100297688","display_name":"Zhitao He","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhitao He","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5119204440","display_name":"Sandeep Polisetty","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sandeep Polisetty","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Zhiyuan Fan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhiyuan Fan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Yuchen Huang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yuchen Huang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100964781","display_name":"Shujin Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shujin Wu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5029408111","display_name":"Yi R. Fung","orcid":"https://orcid.org/0009-0006-1869-0363"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yi R. Fung","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":6,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.7588,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":{"value":0.87855744,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":95,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"16427","last_page":"16444"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10215","display_name":"Semantic Web and Ontologies","score":0.9818999767303467,"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/T10215","display_name":"Semantic Web and Ontologies","score":0.9818999767303467,"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":0.9420999884605408,"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/computer-science","display_name":"Computer science","score":0.6287914514541626},{"id":"https://openalex.org/keywords/calibration","display_name":"Calibration","score":0.6090909838676453},{"id":"https://openalex.org/keywords/boundary","display_name":"Boundary (topology)","score":0.5153266787528992},{"id":"https://openalex.org/keywords/knowledge-management","display_name":"Knowledge management","score":0.33807456493377686},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.14838987588882446},{"id":"https://openalex.org/keywords/statistics","display_name":"Statistics","score":0.11480823159217834}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6287914514541626},{"id":"https://openalex.org/C165838908","wikidata":"https://www.wikidata.org/wiki/Q736777","display_name":"Calibration","level":2,"score":0.6090909838676453},{"id":"https://openalex.org/C62354387","wikidata":"https://www.wikidata.org/wiki/Q875399","display_name":"Boundary (topology)","level":2,"score":0.5153266787528992},{"id":"https://openalex.org/C56739046","wikidata":"https://www.wikidata.org/wiki/Q192060","display_name":"Knowledge management","level":1,"score":0.33807456493377686},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.14838987588882446},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.11480823159217834},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.18653/v1/2025.acl-long.802","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.802","pdf_url":"https://aclanthology.org/2025.acl-long.802.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 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},{"id":"pmh:oai:repository.hkust.edu.hk:1783.1-168524","is_oa":false,"landing_page_url":"http://repository.hkust.edu.hk/ir/Record/1783.1-168524","pdf_url":null,"source":{"id":"https://openalex.org/S4306401796","display_name":"Rare & Special e-Zone (The Hong Kong University of Science and Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I200769079","host_organization_name":"Hong Kong University of Science and Technology","host_organization_lineage":["https://openalex.org/I200769079"],"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":"Conference paper"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.acl-long.802","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.802","pdf_url":"https://aclanthology.org/2025.acl-long.802.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 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[{"display_name":"Reduced inequalities","id":"https://metadata.un.org/sdg/10","score":0.4300000071525574}],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412945065.pdf","grobid_xml":"https://content.openalex.org/works/W4412945065.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/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052"],"abstract_inverted_index":{"In":[0],"recent":[1],"years,":[2],"multimodal":[3,19,174],"large":[4],"language":[5],"models":[6],"(MLLMs)":[7],"have":[8],"made":[9],"significant":[10],"progress":[11],"but":[12,51],"continue":[13],"to":[14,41,53,61,90,98,110,180],"face":[15],"inherent":[16],"challenges":[17],"in":[18,56,173,183],"reasoning,":[20],"which":[21],"requires":[22],"multi-level":[23],"(e.g.,":[24,29],"perception,":[25],"reasoning)":[26],"and":[27,49,94,142,165,178],"multi-granular":[28],"multi-step":[30],"reasoning":[31,58,82,107,147,150],"chain)":[32],"advanced":[33],"inferencing.Prior":[34],"work":[35],"on":[36,43,114],"estimating":[37],"model":[38,140],"confidence":[39,55,84,100,119,124,144,175],"tends":[40],"focus":[42],"the":[44,75,105],"overall":[45],"response":[46],"for":[47,122,137],"training":[48],"calibration,":[50],"fails":[52],"assess":[54],"each":[57,102,146],"step,":[59,148],"leading":[60],"undesirable":[62],"hallucination":[63],"snowballing.In":[64],"this":[65,115],"work,":[66],"we":[67,88,127],"present":[68],"MMBoundary,":[69],"a":[70,129],"novel":[71],"framework":[72],"that":[73,155],"advances":[74],"knowledge":[76,141],"boundary":[77],"awareness":[78],"of":[79,104,117,170],"MLLMs":[80],"through":[81],"step":[83,103],"calibration.To":[85],"achieve":[86],"this,":[87],"propose":[89],"incorporate":[91],"complementary":[92],"textual":[93],"cross-modal":[95],"self-rewarding":[96,118],"signals":[97],"estimate":[99],"at":[101,145],"MLLM":[106,113],"process.In":[108],"addition":[109],"supervised":[111],"fine-tuning":[112],"set":[116],"estimation":[120],"signal":[121],"initial":[123],"expression":[125],"warm-up,":[126],"introduce":[128],"reinforcement":[130],"learning":[131],"stage":[132],"with":[133],"multiple":[134],"reward":[135],"functions":[136],"further":[138],"aligning":[139],"calibrating":[143],"enhancing":[149],"chain":[151],"self-correction.Empirical":[152],"results":[153],"show":[154],"MMBoundary":[156],"significantly":[157],"outperforms":[158],"existing":[159],"methods":[160],"across":[161],"diverse":[162],"domain":[163],"datasets":[164],"metrics,":[166],"achieving":[167],"an":[168],"average":[169],"7.5%":[171],"reduction":[172],"calibration":[176],"errors":[177],"up":[179],"8.3%":[181],"improvement":[182],"task":[184],"performance":[185],"1":[186],".":[187]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}