{"id":"https://openalex.org/W7166871853","doi":"https://doi.org/10.18653/v1/2026.findings-acl.1818","title":"Switching Heads and Softening Tokens: Turnkey Solutions to Visually Grounded Document QA","display_name":"Switching Heads and Softening Tokens: Turnkey Solutions to Visually Grounded Document QA","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7166871853","doi":"https://doi.org/10.18653/v1/2026.findings-acl.1818"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.findings-acl.1818","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1818","pdf_url":"https://aclanthology.org/2026.findings-acl.1818.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 2026","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.findings-acl.1818.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5000482403","display_name":"Ximing Wen","orcid":null},"institutions":[{"id":"https://openalex.org/I2250650973","display_name":"Samsung (South Korea)","ror":"https://ror.org/04w3jy968","country_code":"KR","type":"company","lineage":["https://openalex.org/I2250650973"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Ximing Wen","raw_affiliation_strings":["AI Center-Mountain View , Samsung Electronics Drexel University. AI Center-Mountain View , Samsung Electronics"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"AI Center-Mountain View , Samsung Electronics Drexel University. AI Center-Mountain View , Samsung Electronics","institution_ids":["https://openalex.org/I2250650973"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139841039","display_name":"Wenbo Li","orcid":null},"institutions":[{"id":"https://openalex.org/I2250650973","display_name":"Samsung (South Korea)","ror":"https://ror.org/04w3jy968","country_code":"KR","type":"company","lineage":["https://openalex.org/I2250650973"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Wenbo Li","raw_affiliation_strings":["AI Center-Mountain View , Samsung Electronics Drexel University. AI Center-Mountain View , Samsung Electronics"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"AI Center-Mountain View , Samsung Electronics Drexel University. AI Center-Mountain View , Samsung Electronics","institution_ids":["https://openalex.org/I2250650973"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5062478295","display_name":"Sudipta Paul","orcid":"https://orcid.org/0000-0002-3066-1745"},"institutions":[{"id":"https://openalex.org/I2250650973","display_name":"Samsung (South Korea)","ror":"https://ror.org/04w3jy968","country_code":"KR","type":"company","lineage":["https://openalex.org/I2250650973"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Sudipta Paul","raw_affiliation_strings":["AI Center-Mountain View , Samsung Electronics Drexel University. AI Center-Mountain View , Samsung Electronics"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"AI Center-Mountain View , Samsung Electronics Drexel University. AI Center-Mountain View , Samsung Electronics","institution_ids":["https://openalex.org/I2250650973"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5040083798","display_name":"Yashas Malur Saidutta","orcid":"https://orcid.org/0000-0002-8487-0219"},"institutions":[{"id":"https://openalex.org/I2250650973","display_name":"Samsung (South Korea)","ror":"https://ror.org/04w3jy968","country_code":"KR","type":"company","lineage":["https://openalex.org/I2250650973"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Yashas Malur Saidutta","raw_affiliation_strings":["AI Center-Mountain View , Samsung Electronics Drexel University. AI Center-Mountain View , Samsung Electronics"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"AI Center-Mountain View , Samsung Electronics Drexel University. AI Center-Mountain View , Samsung Electronics","institution_ids":["https://openalex.org/I2250650973"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066326081","display_name":"Kalpa Gunaratna","orcid":"https://orcid.org/0000-0002-9058-4577"},"institutions":[{"id":"https://openalex.org/I2250650973","display_name":"Samsung (South Korea)","ror":"https://ror.org/04w3jy968","country_code":"KR","type":"company","lineage":["https://openalex.org/I2250650973"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Kalpa Gunaratna","raw_affiliation_strings":["AI Center-Mountain View , Samsung Electronics Drexel University. AI Center-Mountain View , Samsung Electronics"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"AI Center-Mountain View , Samsung Electronics Drexel University. AI Center-Mountain View , Samsung Electronics","institution_ids":["https://openalex.org/I2250650973"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5139765618","display_name":"Srinivas Chappidi","orcid":null},"institutions":[{"id":"https://openalex.org/I2250650973","display_name":"Samsung (South Korea)","ror":"https://ror.org/04w3jy968","country_code":"KR","type":"company","lineage":["https://openalex.org/I2250650973"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Srinivas Chappidi","raw_affiliation_strings":["AI Center-Mountain View , Samsung Electronics Drexel University. AI Center-Mountain View , Samsung Electronics"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"AI Center-Mountain View , Samsung Electronics Drexel University. AI Center-Mountain View , Samsung Electronics","institution_ids":["https://openalex.org/I2250650973"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I2250650973"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.94564644,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"36490","last_page":"36503"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12377","display_name":"Digital Humanities and Scholarship","score":0.20720000565052032,"subfield":{"id":"https://openalex.org/subfields/1208","display_name":"Literature and Literary Theory"},"field":{"id":"https://openalex.org/fields/12","display_name":"Arts and Humanities"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T12377","display_name":"Digital Humanities and Scholarship","score":0.20720000565052032,"subfield":{"id":"https://openalex.org/subfields/1208","display_name":"Literature and Literary Theory"},"field":{"id":"https://openalex.org/fields/12","display_name":"Arts and Humanities"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T10799","display_name":"Data Visualization and Analytics","score":0.1898999959230423,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.12630000710487366,"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/turnkey","display_name":"Turnkey","score":0.8069000244140625},{"id":"https://openalex.org/keywords/softening","display_name":"Softening","score":0.4564000070095062},{"id":"https://openalex.org/keywords/work","display_name":"Work (physics)","score":0.30959999561309814},{"id":"https://openalex.org/keywords/materials-testing","display_name":"Materials testing","score":0.25540000200271606}],"concepts":[{"id":"https://openalex.org/C2777843530","wikidata":"https://www.wikidata.org/wiki/Q1151244","display_name":"Turnkey","level":2,"score":0.8069000244140625},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.54339998960495},{"id":"https://openalex.org/C199639397","wikidata":"https://www.wikidata.org/wiki/Q1788588","display_name":"Engineering drawing","level":1,"score":0.47850000858306885},{"id":"https://openalex.org/C2781098093","wikidata":"https://www.wikidata.org/wiki/Q85787439","display_name":"Softening","level":2,"score":0.4564000070095062},{"id":"https://openalex.org/C78519656","wikidata":"https://www.wikidata.org/wiki/Q101333","display_name":"Mechanical engineering","level":1,"score":0.40689998865127563},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.3278000056743622},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.30959999561309814},{"id":"https://openalex.org/C66938386","wikidata":"https://www.wikidata.org/wiki/Q633538","display_name":"Structural engineering","level":1,"score":0.30079999566078186},{"id":"https://openalex.org/C77595967","wikidata":"https://www.wikidata.org/wiki/Q3151013","display_name":"Forensic engineering","level":1,"score":0.26840001344680786},{"id":"https://openalex.org/C2988617437","wikidata":"https://www.wikidata.org/wiki/Q6627318","display_name":"Materials testing","level":2,"score":0.25540000200271606},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.2515000104904175}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.findings-acl.1818","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1818","pdf_url":"https://aclanthology.org/2026.findings-acl.1818.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 2026","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.findings-acl.1818","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1818","pdf_url":"https://aclanthology.org/2026.findings-acl.1818.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 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166871853.pdf","grobid_xml":"https://content.openalex.org/works/W7166871853.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Visually":[0],"Grounded":[1],"Document":[2],"Question":[3],"Answering":[4],"often":[5],"lacks":[6],"robust,":[7,50],"end-to-end":[8],"solutions":[9],"capable":[10],"of":[11,57,122,138,155],"handling":[12],"complex,":[13],"multianswer":[14],"queries":[15],"without":[16],"reliance":[17],"on":[18],"ad-hoc":[19],"processing.In":[20],"this":[21,31,51,78],"work,":[22],"we":[23,63,90,143],"propose":[24,64,91,144],"two":[25],"turnkey":[26],"LLM":[27],"architectures":[28],"to":[29,60],"address":[30,61],"gap.We":[32],"first":[33],"introduce":[34],"a":[35,65,92,95,114],"single-head":[36],"architecture":[37,97],"where":[38],"coordinates":[39],"are":[40],"represented":[41],"as":[42],"special":[43],"tokens":[44],"within":[45],"the":[46,55,83,128,135,139],"unified":[47],"vocabulary.While":[48],"structurally":[49],"approach":[52],"suffers":[53],"from":[54],"limitations":[56],"discrete":[58],"supervision;":[59],"this,":[62],"novel":[66],"\"softening":[67],"token\"":[68],"method":[69,108,147],"that":[70,98,148],"enables":[71],"differentiable":[72],"Mean-Squared-Error":[73],"loss":[74],"over":[75],"token":[76],"probabilities.Although":[77],"significantly":[79],"improves":[80],"visual":[81],"grounding,":[82],"spatial":[84,111],"precision":[85,112,137],"remains":[86],"bound":[87],"by":[88,119,126],"discretization.Consequently,":[89],"second":[93],"solution:":[94],"dual-head":[96],"alternates":[99],"between":[100],"text":[101],"generation":[102],"and":[103],"regression-based":[104],"bounding":[105],"box":[106],"prediction.This":[107],"offers":[109],"high":[110,136],"via":[113],"regression":[115],"head,":[116],"further":[117],"stabilized":[118],"our":[120],"introduction":[121],"an":[123,145],"Intersection-over-Union":[124],"loss.Finally,":[125],"combining":[127],"single":[129],"head":[130,141],"model's":[131],"structural":[132],"robustness":[133],"with":[134],"dual":[140],"model,":[142],"ensemble":[146],"yields":[149],"significant":[150],"performance":[151],"gains":[152],"beyond":[153],"each":[154],"individual":[156],"components.":[157]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-07-02T00:00:00"}