{"id":"https://openalex.org/W7168147670","doi":"https://doi.org/10.48550/arxiv.2607.09623","title":"Task-Specific Multimodal Question Answering Agents via Confidence Calibration and Incremental Reasoning for QANTA 2026","display_name":"Task-Specific Multimodal Question Answering Agents via Confidence Calibration and Incremental Reasoning for QANTA 2026","publication_year":2026,"publication_date":"2026-07-10","ids":{"openalex":"https://openalex.org/W7168147670","doi":"https://doi.org/10.48550/arxiv.2607.09623"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2607.09623","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.09623","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.2607.09623","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5140543649","display_name":"Nirjhar Das","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Das, Nirjhar","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5092780333","display_name":"Md. Al-Mamun Provath","orcid":"https://orcid.org/0000-0002-1203-3331"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Provath, Md. Al-Mamun","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/T10028","display_name":"Topic Modeling","score":0.5788000226020813,"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":0.5788000226020813,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.36800000071525574,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T11273","display_name":"Advanced Graph Neural Networks","score":0.010300000198185444,"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/question-answering","display_name":"Question answering","score":0.809499979019165},{"id":"https://openalex.org/keywords/calibration","display_name":"Calibration","score":0.5527999997138977},{"id":"https://openalex.org/keywords/baseline","display_name":"Baseline (sea)","score":0.5198000073432922},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.5102999806404114},{"id":"https://openalex.org/keywords/selection","display_name":"Selection (genetic algorithm)","score":0.42239999771118164}],"concepts":[{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.809499979019165},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.741599977016449},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6258999705314636},{"id":"https://openalex.org/C165838908","wikidata":"https://www.wikidata.org/wiki/Q736777","display_name":"Calibration","level":2,"score":0.5527999997138977},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.5198000073432922},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.5102999806404114},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.42590001225471497},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.42239999771118164},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4032000005245209},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.3549000024795532},{"id":"https://openalex.org/C91306197","wikidata":"https://www.wikidata.org/wiki/Q45767","display_name":"Competition (biology)","level":2,"score":0.2784000039100647},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.2669999897480011},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.2621999979019165}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2607.09623","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.09623","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.2607.09623","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.09623","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":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"We":[0],"present":[1],"our":[2,149],"submission":[3],"to":[4,56,90,121,134],"the":[5,11,94,165],"QANTA":[6],"2026":[7,13],"shared":[8],"challenge":[9,44],"at":[10],"ICML":[12],"Workshop":[14],"on":[15,142,197],"Efficient":[16],"Multimodal":[17],"Question":[18],"Answering":[19],"(EMM-QA).":[20],"Quanta":[21],"evaluates":[22],"multimodal":[23,131,199],"quizbowl":[24],"systems":[25],"that":[26,106,188],"answer":[27,57,66,137],"pyramid-style":[28],"questions":[29],"from":[30,110],"incrementally":[31],"revealed":[32],"text":[33],"and":[34,60,68,100,130,155,178],"accompanying":[35],"images":[36],"while":[37],"operating":[38],"under":[39,58],"realistic":[40],"efficiency":[41],"constraints.":[42],"The":[43,185],"consists":[45],"of":[46,170,176,183],"two":[47],"distinct":[48],"tasks:":[49],"Tossup":[50,83,174],"questions,":[51,62],"which":[52,63],"require":[53],"deciding":[54],"when":[55],"uncertainty,":[59],"Bonus":[61,115,180],"emphasize":[64],"accurate":[65],"selection":[67],"human":[69],"adoption.":[70],"To":[71],"address":[72],"these":[73],"differing":[74],"objectives,":[75],"we":[76],"develop":[77],"a":[78,86,101,143,159,173,179],"task-specific":[79,190],"two-agent":[80],"architecture.":[81],"Our":[82,114,162],"agent":[84,116],"utilizes":[85],"GPT-4o-mini-class":[87],"model":[88,119,147],"(referred":[89,120],"as":[91,122],"GPT-4.1-mini":[92],"in":[93],"competition":[95],"logs)":[96],"with":[97,124],"confidence-calibrated":[98],"answering":[99,201],"domain-specific":[102],"numeric":[103],"reasoning":[104,153,191],"policy":[105],"reduces":[107],"overconfident":[108],"predictions":[109],"isolated":[111],"quantitative":[112],"clues.":[113],"uses":[117],"GPT-4o-class":[118],"GPT-4.1)":[123],"leadin-aware":[125],"reasoning,":[126,129],"structured":[127],"relational":[128],"evidence":[132],"integration":[133],"improve":[135],"exact":[136],"selection.":[138],"Rather":[139],"than":[140],"relying":[141],"retrieval":[144],"pipeline":[145],"or":[146],"ensembles,":[148],"approach":[150],"emphasizes":[151],"efficient":[152],"policies":[154],"confidence":[156],"calibration":[157],"within":[158],"hosted-only":[160],"environment.":[161],"system":[163],"achieved":[164],"highest":[166],"overall":[167],"leaderboard":[168],"score":[169,175,182],"0.402,":[171],"including":[172],"0.238":[177],"Effect":[181],"0.164.":[184],"results":[186],"demonstrate":[187],"lightweight,":[189],"strategies":[192],"can":[193],"provide":[194],"strong":[195],"performance":[196],"resource-constrained":[198],"question":[200],"benchmarks.":[202]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-07-14T00:00:00"}