{"id":"https://openalex.org/W7224323931","doi":"https://doi.org/10.18653/2026.emnlp-industry.90","title":"Retrieval-Conditional Parsing Score (RCPS): Choosing Document Parsers by Retrieval, Not by Appearance","display_name":"Retrieval-Conditional Parsing Score (RCPS): Choosing Document Parsers by Retrieval, Not by Appearance","alternate_titles":[],"publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7224323931","doi":"https://doi.org/10.18653/2026.emnlp-industry.90"},"language":"en","primary_location":{"id":"doi:10.18653/2026.emnlp-industry.90","doi":"https://doi.org/10.18653/2026.emnlp-industry.90","is_oa":true,"landing_page_url":"https://doi.org/10.18653/2026.emnlp-industry.90","pdf_url":"https://aclanthology.org/2026.emnlp-industry.90.pdf","source":{"id":"https://openalex.org/S4363608991","display_name":"Conference on Empirical Methods in Natural Language Processing (EMNLP)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/P4310320244","host_organization_name":"Association for Computational Linguistics","host_organization_lineage":["https://openalex.org/P4310320244"],"host_organization_lineage_names":["Association for Computational Linguistics"],"type":"conference"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track","raw_type":"proceedings-article"},"parent_works":[],"lineage":["https://openalex.org/W7224323931"],"child_works_count":0,"type":"conference-paper","is_traditional":true,"indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://aclanthology.org/2026.emnlp-industry.90.pdf","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5139759789","display_name":"Sang-Woo Son","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sang-Woo Son","raw_affiliation_strings":["WIGTN , Seoul , Republic of Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"WIGTN , Seoul , Republic of Korea","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5022559416","display_name":"Hyeong\u2010Seob Kim","orcid":"https://orcid.org/0000-0002-0964-0288","observed_orcids":["https://orcid.org/0000-0002-0964-0288"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hyeong-seob Kim","raw_affiliation_strings":["WIGTN , Seoul , Republic of Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"WIGTN , Seoul , Republic of Korea","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139749094","display_name":"Hyeonsang Kim","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hyeonsang Kim","raw_affiliation_strings":["WIGTN , Seoul , Republic of Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"WIGTN , Seoul , Republic of Korea","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5160126027","display_name":"Hyun-woo Cho","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hyun-woo Cho","raw_affiliation_strings":["WIGTN , Seoul , Republic of Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"WIGTN , Seoul , Republic of Korea","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101891243","display_name":"Jinmo Kim","orcid":"https://orcid.org/0000-0001-8614-917X","observed_orcids":["https://orcid.org/0000-0001-8614-917X"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jinmo Kim","raw_affiliation_strings":["WIGTN , Seoul , Republic of Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"WIGTN , Seoul , Republic of Korea","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.0,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1531","last_page":"1544"},"is_retracted":false,"is_pulled":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10286","display_name":"Information Retrieval and Search Behavior","score":0.5112000107765198,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T10286","display_name":"Information Retrieval and Search Behavior","score":0.5112000107765198,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T10028","display_name":"Topic Modeling","score":0.38429999351501465,"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":0.04270000010728836,"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/document-parsing","display_name":"document parsing","score":0.6359999775886536},{"id":"https://openalex.org/keywords/mean-reciprocal-rank","display_name":"mean reciprocal rank","score":0.7099999785423279},{"id":"https://openalex.org/keywords/table-parsing","display_name":"table parsing","score":0.4410000145435333},{"id":"https://openalex.org/keywords/preference-optimization","display_name":"preference optimization","score":0.5820000171661377}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7354000210762024},{"id":"https://openalex.org/C186644900","wikidata":"https://www.wikidata.org/wiki/Q194152","display_name":"Parsing","level":2,"score":0.6740000247955322},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5972999930381775},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.583899974822998},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.32199999690055847},{"id":"https://openalex.org/C60690694","wikidata":"https://www.wikidata.org/wiki/Q894902","display_name":"Bottom-up parsing","level":4,"score":0.2669000029563904},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.24950000643730164},{"id":"https://openalex.org/C148524875","wikidata":"https://www.wikidata.org/wiki/Q6975395","display_name":"F1 score","level":2,"score":0.22619999945163727},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2152000069618225},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.21369999647140503}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/2026.emnlp-industry.90","doi":"https://doi.org/10.18653/2026.emnlp-industry.90","is_oa":true,"landing_page_url":"https://doi.org/10.18653/2026.emnlp-industry.90","pdf_url":"https://aclanthology.org/2026.emnlp-industry.90.pdf","source":{"id":"https://openalex.org/S4363608991","display_name":"Conference on Empirical Methods in Natural Language Processing (EMNLP)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/P4310320244","host_organization_name":"Association for Computational Linguistics","host_organization_lineage":["https://openalex.org/P4310320244"],"host_organization_lineage_names":["Association for Computational Linguistics"],"type":"conference"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/2026.emnlp-industry.90","doi":"https://doi.org/10.18653/2026.emnlp-industry.90","is_oa":true,"landing_page_url":"https://doi.org/10.18653/2026.emnlp-industry.90","pdf_url":"https://aclanthology.org/2026.emnlp-industry.90.pdf","source":{"id":"https://openalex.org/S4363608991","display_name":"Conference on Empirical Methods in Natural Language Processing (EMNLP)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/P4310320244","host_organization_name":"Association for Computational Linguistics","host_organization_lineage":["https://openalex.org/P4310320244"],"host_organization_lineage_names":["Association for Computational Linguistics"],"type":"conference"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track","raw_type":"proceedings-article"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7224323931.pdf","grobid_xml":"https://content.openalex.org/works/W7224323931.grobid-xml"},"referenced_works_count":24,"referenced_works":["https://openalex.org/W4391709827","https://openalex.org/W4402670856","https://openalex.org/W4403583523","https://openalex.org/W4403882995","https://openalex.org/W4403928906","https://openalex.org/W4404342118","https://openalex.org/W4404781515","https://openalex.org/W4405057528","https://openalex.org/W4412945697","https://openalex.org/W4412945731","https://openalex.org/W4413145313","https://openalex.org/W4413733698","https://openalex.org/W4415191111","https://openalex.org/W4415795501","https://openalex.org/W4416035839","https://openalex.org/W4416036830","https://openalex.org/W4416780398","https://openalex.org/W4417252869","https://openalex.org/W7133208539","https://openalex.org/W7133243847","https://openalex.org/W7158994651","https://openalex.org/W7166803669","https://openalex.org/W7166845830","https://openalex.org/W7166879473"],"related_works":["https://openalex.org/W4385567066","https://openalex.org/W7131831242","https://openalex.org/W3112013081","https://openalex.org/W2573278614","https://openalex.org/W7096679939","https://openalex.org/W3029510630","https://openalex.org/W2483518559","https://openalex.org/W7065546348","https://openalex.org/W1605120687","https://openalex.org/W2094563091"],"abstract_inverted_index":{"Document":[0],"RAG":[1],"retrieves":[2],"from":[3,102],"parser":[4],"outputs,":[5],"yet":[6],"parsers":[7,94],"are":[8],"commonly":[9],"selected":[10],"by":[11,22,164],"intrinsic":[12],"measures":[13],"of":[14,126],"transcription":[15],"fidelity":[16],"and":[17,53,114,167,192,202],"boundary":[18],"quality":[19],"rather":[20],"than":[21,75,136],"retrieval":[23,43,54,144],"performance.We":[24],"introduce":[25],"the":[26,57,108,198],"Retrieval-Conditional":[27],"Parsing":[28],"Score":[29],"(RCPS),":[30],"a":[31,39,70,153,178],"trainingfree":[32],"protocol":[33],"that":[34],"ranks":[35],"parser-chunker":[36],"combinations":[37],"on":[38,96,173],"fixed":[40],"held-out":[41],"Q-A":[42],"probe.RCPS":[44],"averages":[45],"mean":[46],"reciprocal":[47],"rank":[48],"(MRR)":[49],"across":[50],"specified":[51],"retrievers":[52],"depths,":[55],"using":[56],"same":[58],"format-normalised":[59],"exact-span":[60],"relevance":[61],"rule":[62],"for":[63,124],"every":[64],"candidate.On":[65],"KoGovDoc-RAG,":[66],"table-enabled":[67],"MinerUon":[68],"has":[69],"higher":[71],"Boundary":[72,146],"Clarity":[73,147],"score":[74,111],"our":[76],"production":[77],"parser,":[78],"Prod":[79,106,130,172],"(0.713":[80],"versus":[81,91],"0.610).Yet":[82],"its":[83],"Hit@1":[84],"is":[85],"42.6":[86],"percentage":[87,169],"points":[88,170],"lower":[89],"(0.123":[90],"0.549).Across":[92],"five":[93],"evaluated":[95],"all":[97],"294":[98],"pages,":[99],"RCPS":[100,204,210],"ranges":[101],"0.137":[103],"to":[104],"0.584.With":[105],"fixed,":[107],"four":[109],"chunkers":[110],"between":[112],"0.535":[113],"0.593.A":[115],"retriever-free":[116],"coverage":[117],"diagnostic":[118],"finds":[119],"no":[120],"normalised":[121],"exact":[122],"match":[123],"20.2%":[125],"reference":[127,205],"spans":[128],"in":[129],"output.No":[131],"tested":[132],"chunker":[133],"splits":[134],"more":[135],"2.3%.On":[137],"sourcealigned":[138],"OHR-Bench":[139],"data,":[140],"semantic":[141],"noise":[142],"lowers":[143],"while":[145],"remains":[148],"stable":[149],"or":[150],"changes":[151],"non-monotonically.In":[152],"secondary":[154],"parser-training":[155],"study,":[156],"two":[157],"direct":[158],"preference":[159],"optimization":[160],"checkpoints":[161],"improve":[162],"Hit@5":[163],"only":[165,194],"0.95":[166],"1.15":[168],"over":[171],"an":[174,203],"audited":[175],"compatibility":[176],"subset;":[177],"matched":[179],"edit-distance":[180],"control":[181],"performs":[182],"similarly.The":[183],"workflow":[184],"therefore":[185],"selects":[186],"with":[187,190],"RCPS,":[188],"diagnoses":[189],"coverage,":[191],"trains":[193],"if":[195],"needed.We":[196],"release":[197],"663-Q-A":[199],"KoGovDoc-RAG":[200],"probe":[201],"implementation.Evaluation":[206],"frame":[207],"Candidate":[208],"pipelines":[209],"evaluation":[211],"Select,":[212],"diagnose":[213],"+":[214],"act":[215]},"cited_by_api_url":"https://api.openalex.org/works?filter=cites:W7224323931","counts_by_year":[],"updated_date":"2026-11-17T12:49:18.490984","created_date":"2026-10-30T00:00:00"}
