{"id":"https://openalex.org/W7131072084","doi":"https://doi.org/10.48550/arxiv.2602.17687","title":"IRPAPERS: A Visual Document Benchmark for Scientific Retrieval and Question Answering","display_name":"IRPAPERS: A Visual Document Benchmark for Scientific Retrieval and Question Answering","publication_year":2026,"publication_date":"2026-02-05","ids":{"openalex":"https://openalex.org/W7131072084","doi":"https://doi.org/10.48550/arxiv.2602.17687"},"language":null,"primary_location":{"id":"pmh:doi:10.48550/arxiv.2602.17687","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":null,"any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5051577117","display_name":"Connor Shorten","orcid":"https://orcid.org/0000-0001-6253-6861"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Shorten, Connor","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5126608446","display_name":"Augustas Skaburskas","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Skaburskas, Augustas","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5126622042","display_name":"Daniel M. Jones","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jones, Daniel M.","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5115458017","display_name":"Charles Pierse","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pierse, Charles","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5126595284","display_name":"Roberto Esposito","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Esposito, Roberto","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5115458018","display_name":"John Trengrove","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Trengrove, John","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5126596144","display_name":"Etienne Dilocker","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dilocker, Etienne","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5047498413","display_name":"Bob van Luijt","orcid":"https://orcid.org/0000-0002-8239-7421"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"van Luijt, Bob","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":8,"corresponding_author_ids":["https://openalex.org/A5051577117"],"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9714000225067139,"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"}},"topics":[{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9714000225067139,"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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.008100000210106373,"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/T10028","display_name":"Topic Modeling","score":0.00559999980032444,"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":0.7195000052452087},{"id":"https://openalex.org/keywords/document-retrieval","display_name":"Document retrieval","score":0.5654000043869019},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.5425999760627747},{"id":"https://openalex.org/keywords/image-retrieval","display_name":"Image retrieval","score":0.4772000014781952},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.4575999975204468},{"id":"https://openalex.org/keywords/embedding","display_name":"Embedding","score":0.44850000739097595},{"id":"https://openalex.org/keywords/visual-word","display_name":"Visual Word","score":0.4259999990463257},{"id":"https://openalex.org/keywords/relevance","display_name":"Relevance (law)","score":0.40369999408721924}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7980999946594238},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.7976999878883362},{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.7195000052452087},{"id":"https://openalex.org/C161156560","wikidata":"https://www.wikidata.org/wiki/Q1638872","display_name":"Document retrieval","level":2,"score":0.5654000043869019},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5425999760627747},{"id":"https://openalex.org/C1667742","wikidata":"https://www.wikidata.org/wiki/Q10927554","display_name":"Image retrieval","level":3,"score":0.4772000014781952},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.4575999975204468},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.44850000739097595},{"id":"https://openalex.org/C189391414","wikidata":"https://www.wikidata.org/wiki/Q7936579","display_name":"Visual Word","level":4,"score":0.4259999990463257},{"id":"https://openalex.org/C158154518","wikidata":"https://www.wikidata.org/wiki/Q7310970","display_name":"Relevance (law)","level":2,"score":0.40369999408721924},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.39809998869895935},{"id":"https://openalex.org/C55166926","wikidata":"https://www.wikidata.org/wiki/Q2892946","display_name":"Oracle","level":2,"score":0.3610999882221222},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.336899995803833},{"id":"https://openalex.org/C75165309","wikidata":"https://www.wikidata.org/wiki/Q2258979","display_name":"Search engine indexing","level":2,"score":0.3167000114917755},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.3059999942779541},{"id":"https://openalex.org/C177937566","wikidata":"https://www.wikidata.org/wiki/Q4223102","display_name":"Document clustering","level":3,"score":0.30230000615119934},{"id":"https://openalex.org/C2780226545","wikidata":"https://www.wikidata.org/wiki/Q6888030","display_name":"Modality (human\u2013computer interaction)","level":2,"score":0.2791000008583069},{"id":"https://openalex.org/C20228898","wikidata":"https://www.wikidata.org/wiki/Q83540","display_name":"Full text search","level":3,"score":0.27320000529289246},{"id":"https://openalex.org/C2985933255","wikidata":"https://www.wikidata.org/wiki/Q1638872","display_name":"Text retrieval","level":2,"score":0.2712000012397766},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.26579999923706055},{"id":"https://openalex.org/C179926584","wikidata":"https://www.wikidata.org/wiki/Q207714","display_name":"Transcription (linguistics)","level":2,"score":0.260699987411499},{"id":"https://openalex.org/C130590232","wikidata":"https://www.wikidata.org/wiki/Q1671754","display_name":"Inverted index","level":3,"score":0.25049999356269836}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:doi:10.48550/arxiv.2602.17687","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},{"id":"doi:10.48550/arxiv.2602.17687","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.17687","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"pmh:doi:10.48550/arxiv.2602.17687","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},"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":{"AI":[0],"systems":[1,21,56,196,203],"have":[2],"achieved":[3,110,197],"remarkable":[4],"success":[5],"in":[6,36],"processing":[7,15],"text":[8,31,108,176,229],"and":[9,32,42,78,92,95,106,115,124,146,156,178,188,207,230,233,247],"relational":[10],"data,":[11],"yet":[12],"visual":[13,28],"document":[14,46],"remains":[16],"relatively":[17],"underexplored.":[18],"Whereas":[19],"traditional":[20],"require":[22,238],"OCR":[23,80],"transcriptions":[24],"to":[25,58,137],"convert":[26],"these":[27],"documents":[29],"into":[30],"metadata,":[33],"recent":[34],"advances":[35],"multimodal":[37,134],"foundation":[38],"models":[39],"offer":[40],"retrieval":[41,94,100,120,213,217],"generation":[43],"directly":[44],"from":[45,70,211],"images.":[47],"This":[48],"raises":[49],"a":[50,65],"key":[51],"question:":[52],"How":[53],"do":[54],"image-based":[55,119,202],"compare":[57,90],"established":[59],"text-based":[60,93,194],"methods?":[61],"We":[62,149,222],"introduce":[63],"IRPAPERS,":[64],"benchmark":[66],"of":[67,227],"3,230":[68],"pages":[69],"166":[71],"scientific":[72],"papers,":[73],"with":[74,154,215],"both":[75,208],"an":[76,79],"image":[77,160,170,231],"transcription":[81],"for":[82],"each":[83],"page.":[84],"Using":[85],"180":[86],"needle-in-the-haystack":[87],"questions,":[88],"we":[89],"image-":[91],"question":[96,192,235],"answering":[97],"systems.":[98],"Text":[99],"using":[101],"Arctic":[102],"2.0":[103],"embeddings,":[104],"BM25,":[105],"hybrid":[107,135],"search":[109,136],"46%":[111],"Recall@1,":[112,143,185],"78%":[113],"Recall@5,":[114,145,187],"91%":[116],"Recall@20,":[117],"while":[118],"reaches":[121],"43%,":[122],"78%,":[123],"93%,":[125],"respectively.":[126],"The":[127,244],"two":[128],"modalities":[129],"exhibit":[130],"complementary":[131,225],"failures,":[132],"enabling":[133],"outperform":[138,172],"either":[139],"alone,":[140],"achieving":[141,183],"49%":[142],"81%":[144],"95%":[147],"Recall@20.":[148,190],"further":[150],"evaluate":[151],"efficiency-performance":[152],"tradeoffs":[153],"MUVERA":[155],"assess":[157],"multiple":[158],"multi-vector":[159],"embedding":[161],"models.":[162],"Among":[163],"closed-source":[164],"models,":[165,182],"Cohere":[166],"Embed":[167],"v4":[168],"page":[169],"embeddings":[171,177],"Voyage":[173],"3":[174],"Large":[175],"all":[179,248],"tested":[180],"open-source":[181],"58%":[184],"87%":[186],"97%":[189],"For":[191],"answering,":[193],"RAG":[195],"higher":[198],"ground-truth":[199],"alignment":[200],"than":[201],"(0.82":[204],"vs.":[205],"0.71),":[206],"benefit":[209],"substantially":[210],"increased":[212],"depth,":[214],"multi-document":[216],"outperforming":[218],"oracle":[219],"single-document":[220],"retrieval.":[221],"analyze":[223],"the":[224,242],"limitations":[226],"unimodal":[228],"representations":[232],"identify":[234],"types":[236],"that":[237],"one":[239],"modality":[240],"over":[241],"other.":[243],"IRPAPERS":[245],"dataset":[246],"experimental":[249],"code":[250],"are":[251],"publicly":[252],"available.":[253]},"counts_by_year":[],"updated_date":"2026-04-04T16:13:02.066488","created_date":"2026-02-24T00:00:00"}