{"id":"https://openalex.org/W7143441520","doi":"https://doi.org/10.48550/arxiv.2603.26049","title":"Seeing Like Radiologists: Context- and Gaze-Guided Vision-Language Pretraining for Chest X-rays","display_name":"Seeing Like Radiologists: Context- and Gaze-Guided Vision-Language Pretraining for Chest X-rays","publication_year":2026,"publication_date":"2026-03-27","ids":{"openalex":"https://openalex.org/W7143441520","doi":"https://doi.org/10.48550/arxiv.2603.26049"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.26049","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.26049","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":null,"license_id":null,"version":null,"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":"https://doi.org/10.48550/arxiv.2603.26049","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5130998252","display_name":"Kang Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Liu, Kang","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130995689","display_name":"Zhuoqi Ma","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ma, Zhuoqi","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130963502","display_name":"Siyu Liang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liang, Siyu","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129745718","display_name":"Yunan Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Yunan","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5023488782","display_name":"Xiyue Gao","orcid":"https://orcid.org/0000-0003-4370-6954"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gao, Xiyue","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130965411","display_name":"Chao Liang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liang, Chao","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130979506","display_name":"Kun Xie","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xie, Kun","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5130963314","display_name":"Qiguang Miao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Miao, Qiguang","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":8,"corresponding_author_ids":["https://openalex.org/A5130998252"],"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":0.6007000207901001,"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":0.6007000207901001,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.15870000422000885,"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/T11775","display_name":"COVID-19 diagnosis using AI","score":0.12700000405311584,"subfield":{"id":"https://openalex.org/subfields/2741","display_name":"Radiology, Nuclear Medicine and Imaging"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.5965999960899353},{"id":"https://openalex.org/keywords/gaze","display_name":"Gaze","score":0.5705000162124634},{"id":"https://openalex.org/keywords/prior-probability","display_name":"Prior probability","score":0.5625},{"id":"https://openalex.org/keywords/salient","display_name":"Salient","score":0.5594000220298767},{"id":"https://openalex.org/keywords/probabilistic-logic","display_name":"Probabilistic logic","score":0.5241000056266785},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.43140000104904175}],"concepts":[{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.5965999960899353},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5924000144004822},{"id":"https://openalex.org/C2779916870","wikidata":"https://www.wikidata.org/wiki/Q14467155","display_name":"Gaze","level":2,"score":0.5705000162124634},{"id":"https://openalex.org/C177769412","wikidata":"https://www.wikidata.org/wiki/Q278090","display_name":"Prior probability","level":3,"score":0.5625},{"id":"https://openalex.org/C2780719617","wikidata":"https://www.wikidata.org/wiki/Q1030752","display_name":"Salient","level":2,"score":0.5594000220298767},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5375000238418579},{"id":"https://openalex.org/C49937458","wikidata":"https://www.wikidata.org/wiki/Q2599292","display_name":"Probabilistic logic","level":2,"score":0.5241000056266785},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.43140000104904175},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.3727000057697296},{"id":"https://openalex.org/C31601959","wikidata":"https://www.wikidata.org/wiki/Q931309","display_name":"Medical imaging","level":2,"score":0.37209999561309814},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3271999955177307},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.3249000012874603},{"id":"https://openalex.org/C56461940","wikidata":"https://www.wikidata.org/wiki/Q970687","display_name":"Eye tracking","level":2,"score":0.290800005197525},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.2791999876499176},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.2782000005245209},{"id":"https://openalex.org/C100776233","wikidata":"https://www.wikidata.org/wiki/Q2532492","display_name":"Bridge (graph theory)","level":2,"score":0.2655999958515167},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2630999982357025}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.26049","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.26049","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2603.26049","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.26049","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":null,"license_id":null,"version":null,"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":{"Despite":[0],"recent":[1],"advances":[2],"in":[3],"medical":[4],"vision-language":[5,63],"pretraining,":[6],"existing":[7,38],"models":[8,77],"still":[9],"struggle":[10],"to":[11,92,132,153],"capture":[12],"the":[13,43],"diagnostic":[14,89,94,117],"workflow:":[15],"radiographs":[16],"are":[17],"typically":[18],"treated":[19],"as":[20,129],"context-agnostic":[21],"images,":[22],"while":[23],"radiologists'":[24,127],"gaze":[25,128],"--":[26,33,83,91],"a":[27,59,72,99],"crucial":[28],"cue":[29],"for":[30,66,159,167,173],"visual":[31],"reasoning":[32],"remains":[34],"largely":[35],"underexplored":[36],"by":[37],"methods.":[39],"These":[40],"limitations":[41],"hinder":[42],"modeling":[44],"of":[45],"disease-specific":[46],"patterns":[47],"and":[48,61,88,107,124,156,161,170],"weaken":[49],"cross-modal":[50,121],"alignment.":[51],"To":[52],"bridge":[53],"this":[54],"gap,":[55],"we":[56],"introduce":[57],"CoGaze,":[58],"Context-":[60],"Gaze-guided":[62],"pretraining":[64],"framework":[65],"chest":[67],"X-rays.":[68],"We":[69,96],"first":[70],"propose":[71],"context-infused":[73],"vision":[74],"encoder":[75],"that":[76,103,142],"how":[78],"radiologists":[79],"integrate":[80],"clinical":[81],"context":[82],"including":[84],"patient":[85],"history,":[86],"symptoms,":[87],"intent":[90],"guide":[93,133],"reasoning.":[95],"then":[97],"present":[98],"multi-level":[100],"supervision":[101],"paradigm":[102],"(1)":[104],"enforces":[105],"intra-":[106],"inter-modal":[108],"semantic":[109],"alignment":[110],"through":[111],"hybrid-positive":[112],"contrastive":[113],"learning,":[114,123],"(2)":[115],"injects":[116],"priors":[118,131],"via":[119],"disease-aware":[120],"representation":[122],"(3)":[125],"leverages":[126],"probabilistic":[130],"attention":[134],"toward":[135],"diagnostically":[136],"salient":[137],"regions.":[138],"Extensive":[139],"experiments":[140],"demonstrate":[141],"CoGaze":[143],"consistently":[144],"outperforms":[145],"state-of-the-art":[146],"methods":[147],"across":[148],"diverse":[149],"tasks,":[150],"achieving":[151],"up":[152],"+2.0%":[154],"CheXbertF1":[155],"+1.2%":[157],"BLEU2":[158],"free-text":[160],"structured":[162],"report":[163],"generation,":[164],"+23.2%":[165],"AUROC":[166],"zero-shot":[168],"classification,":[169],"+12.2%":[171],"Precision@1":[172],"image-text":[174],"retrieval.":[175],"Code":[176],"is":[177],"available":[178],"at":[179],"https://github.com/mk-runner/CoGaze.":[180]},"counts_by_year":[],"updated_date":"2026-03-31T06:07:48.031334","created_date":"2026-03-31T00:00:00"}