{"id":"https://openalex.org/W7148248457","doi":"https://doi.org/10.48550/arxiv.2604.00313","title":"Label-efficient underwater species classification with logistic regression on frozen foundation model embeddings","display_name":"Label-efficient underwater species classification with logistic regression on frozen foundation model embeddings","publication_year":2026,"publication_date":"2026-03-31","ids":{"openalex":"https://openalex.org/W7148248457","doi":"https://doi.org/10.48550/arxiv.2604.00313"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.00313","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.00313","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.2604.00313","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5120751884","display_name":"Thomas Manuel Rost","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Rost, Thomas Manuel","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":1,"corresponding_author_ids":["https://openalex.org/A5120751884"],"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/T11387","display_name":"Ichthyology and Marine Biology","score":0.10679999738931656,"subfield":{"id":"https://openalex.org/subfields/2309","display_name":"Nature and Landscape Conservation"},"field":{"id":"https://openalex.org/fields/23","display_name":"Environmental Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11387","display_name":"Ichthyology and Marine Biology","score":0.10679999738931656,"subfield":{"id":"https://openalex.org/subfields/2309","display_name":"Nature and Landscape Conservation"},"field":{"id":"https://openalex.org/fields/23","display_name":"Environmental Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.09740000218153,"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/T12859","display_name":"Cell Image Analysis Techniques","score":0.07400000095367432,"subfield":{"id":"https://openalex.org/subfields/1304","display_name":"Biophysics"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/discriminative-model","display_name":"Discriminative model","score":0.8457000255584717},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.5877000093460083},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.5546000003814697},{"id":"https://openalex.org/keywords/embedding","display_name":"Embedding","score":0.5317000150680542},{"id":"https://openalex.org/keywords/underwater","display_name":"Underwater","score":0.5192999839782715},{"id":"https://openalex.org/keywords/baseline","display_name":"Baseline (sea)","score":0.49950000643730164},{"id":"https://openalex.org/keywords/transfer-of-learning","display_name":"Transfer of learning","score":0.4837000072002411},{"id":"https://openalex.org/keywords/class","display_name":"Class (philosophy)","score":0.4603999853134155},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.44909998774528503}],"concepts":[{"id":"https://openalex.org/C97931131","wikidata":"https://www.wikidata.org/wiki/Q5282087","display_name":"Discriminative model","level":2,"score":0.8457000255584717},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6176999807357788},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5877000093460083},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.5546000003814697},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.5317000150680542},{"id":"https://openalex.org/C98083399","wikidata":"https://www.wikidata.org/wiki/Q3246517","display_name":"Underwater","level":2,"score":0.5192999839782715},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.49950000643730164},{"id":"https://openalex.org/C150899416","wikidata":"https://www.wikidata.org/wiki/Q1820378","display_name":"Transfer of learning","level":2,"score":0.4837000072002411},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4819999933242798},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.4603999853134155},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.44909998774528503},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.43880000710487366},{"id":"https://openalex.org/C58489278","wikidata":"https://www.wikidata.org/wiki/Q1172284","display_name":"Data set","level":2,"score":0.4334999918937683},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.38499999046325684},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.3756999969482422},{"id":"https://openalex.org/C136389625","wikidata":"https://www.wikidata.org/wiki/Q334384","display_name":"Supervised learning","level":3,"score":0.3546999990940094},{"id":"https://openalex.org/C169903167","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Test set","level":2,"score":0.32510000467300415},{"id":"https://openalex.org/C2776145971","wikidata":"https://www.wikidata.org/wiki/Q30673951","display_name":"Labeled data","level":2,"score":0.320499986410141},{"id":"https://openalex.org/C2780966255","wikidata":"https://www.wikidata.org/wiki/Q5474306","display_name":"Foundation (evidence)","level":2,"score":0.3077000081539154},{"id":"https://openalex.org/C2776321320","wikidata":"https://www.wikidata.org/wiki/Q857525","display_name":"Annotation","level":2,"score":0.287200003862381},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.28439998626708984},{"id":"https://openalex.org/C16910744","wikidata":"https://www.wikidata.org/wiki/Q7705759","display_name":"Test data","level":2,"score":0.28220000863075256},{"id":"https://openalex.org/C3017894346","wikidata":"https://www.wikidata.org/wiki/Q1353040","display_name":"Coast guard","level":2,"score":0.2784000039100647},{"id":"https://openalex.org/C75294576","wikidata":"https://www.wikidata.org/wiki/Q5165192","display_name":"Contextual image classification","level":3,"score":0.26510000228881836},{"id":"https://openalex.org/C62354387","wikidata":"https://www.wikidata.org/wiki/Q875399","display_name":"Boundary (topology)","level":2,"score":0.2572999894618988},{"id":"https://openalex.org/C184898388","wikidata":"https://www.wikidata.org/wiki/Q1435712","display_name":"Pairwise comparison","level":2,"score":0.25369998812675476},{"id":"https://openalex.org/C110083411","wikidata":"https://www.wikidata.org/wiki/Q1744628","display_name":"Statistical classification","level":2,"score":0.25029999017715454}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.00313","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.00313","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.2604.00313","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.00313","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":[{"id":"https://metadata.un.org/sdg/14","score":0.49129486083984375,"display_name":"Life below water"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Automated":[0],"species":[1,130,160],"classification":[2],"from":[3],"underwater":[4,134],"imagery":[5],"is":[6,177],"bottlenecked":[7],"by":[8],"the":[9,59,86,129,133,167],"cost":[10],"of":[11,80,91,105],"expert":[12],"annotation,":[13],"and":[14,56,147,183],"supervised":[15,87,116],"models":[16],"trained":[17],"on":[18,33,58,89,166],"one":[19],"dataset":[20],"rarely":[21],"transfer":[22],"to":[23,75],"new":[24],"conditions.":[25],"We":[26],"investigate":[27],"whether":[28],"a":[29,52,78,152,178],"simple":[30],"classifier":[31,55],"operating":[32],"frozen":[34,43,122],"foundation":[35],"model":[36],"embeddings":[37,46],"can":[38],"close":[39],"this":[40],"gap.":[41],"Using":[42],"DINOv3":[44],"ViT-B/16":[45],"with":[47,97,114],"no":[48,139,143,148],"fine-tuning,":[49],"we":[50],"train":[51],"logistic":[53,68],"regression":[54,69],"evaluate":[57],"AQUA20":[60],"benchmark":[61],"(20":[62],"marine":[63,159],"species).":[64],"At":[65],"full":[66],"supervision,":[67],"achieves":[70],"88.5%":[71],"macro":[72,108],"F1":[73,109],"compared":[74],"ConvNeXt's":[76],"88.9%,":[77],"gap":[79],"0.4":[81],"percentage":[82],"points,":[83],"while":[84],"outperforming":[85],"baseline":[88,156],"8":[90],"20":[92],"species.":[93],"Under":[94],"label":[95],"scarcity,":[96],"21":[98],"labeled":[99],"examples":[100],"per":[101],"class":[102],"(approximately":[103],"6%":[104],"training":[106],"labels),":[107],"exceeds":[110],"80%.":[111],"The":[112],"near-parity":[113],"end-to-end":[115],"learning":[117,141],"demonstrates":[118],"that":[119],"these":[120],"general-purpose,":[121],"representations":[123],"exhibit":[124],"strong":[125],"linear":[126],"separability":[127],"at":[128],"level":[131],"in":[132],"domain.":[135],"Our":[136],"approach":[137],"requires":[138],"deep":[140],"training,":[142],"domain-specific":[144],"data":[145],"engineering,":[146],"underwater-adapted":[149],"models,":[150],"establishing":[151],"practical,":[153],"immediately":[154],"deployable":[155],"for":[157],"label-efficient":[158],"recognition.":[161],"All":[162],"results":[163],"are":[164,185],"reported":[165],"held-out":[168],"test":[169],"set":[170],"over":[171],"100":[172],"random":[173],"seed":[174],"initialisations.":[175],"This":[176],"preliminary":[179],"report;":[180],"further":[181],"evaluations":[182],"ablations":[184],"forthcoming.":[186]},"counts_by_year":[],"updated_date":"2026-04-18T05:59:34.339393","created_date":"2026-04-03T00:00:00"}