{"id":"https://openalex.org/W7135173526","doi":"https://doi.org/10.1109/iccp68926.2025.11427128","title":"CAPSD: Confidence-Aware Promptable Segmentation Distillation for Small Dataset Transformer-Based Models","display_name":"CAPSD: Confidence-Aware Promptable Segmentation Distillation for Small Dataset Transformer-Based Models","publication_year":2025,"publication_date":"2025-10-16","ids":{"openalex":"https://openalex.org/W7135173526","doi":"https://doi.org/10.1109/iccp68926.2025.11427128"},"language":null,"primary_location":{"id":"doi:10.1109/iccp68926.2025.11427128","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccp68926.2025.11427128","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE 21st International Conference on Intelligent Computer Communication and Processing (ICCP)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5128922072","display_name":"Andreea-Maria Onaci","orcid":null},"institutions":[{"id":"https://openalex.org/I4210127226","display_name":"Robert Bosch (Slovenia)","ror":"https://ror.org/03248dr08","country_code":"SI","type":"company","lineage":["https://openalex.org/I4210127226","https://openalex.org/I889804353"]}],"countries":["SI"],"is_corresponding":true,"raw_author_name":"Andreea-Maria Onaci","raw_affiliation_strings":["Robert Bosch SRL,Engineering Systems Vision,Cluj-Napoca,Romania"],"affiliations":[{"raw_affiliation_string":"Robert Bosch SRL,Engineering Systems Vision,Cluj-Napoca,Romania","institution_ids":["https://openalex.org/I4210127226"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5016301087","display_name":"Mihai Negru","orcid":"https://orcid.org/0000-0001-8039-8415"},"institutions":[{"id":"https://openalex.org/I158333966","display_name":"Technical University of Cluj-Napoca","ror":"https://ror.org/03r8nwp71","country_code":"RO","type":"education","lineage":["https://openalex.org/I158333966"]}],"countries":["RO"],"is_corresponding":false,"raw_author_name":"Mihai Negru","raw_affiliation_strings":["Technical University of Cluj-Napoca,Computer Science Department,Cluj-Napoca,Romania"],"affiliations":[{"raw_affiliation_string":"Technical University of Cluj-Napoca,Computer Science Department,Cluj-Napoca,Romania","institution_ids":["https://openalex.org/I158333966"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5128922072"],"corresponding_institution_ids":["https://openalex.org/I4210127226"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.76726151,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"8"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.19169999659061432,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.19169999659061432,"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/T11512","display_name":"Anomaly Detection Techniques and Applications","score":0.0982000008225441,"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/T12535","display_name":"Machine Learning and Data Classification","score":0.07829999923706055,"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/segmentation","display_name":"Segmentation","score":0.46149998903274536},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.43140000104904175},{"id":"https://openalex.org/keywords/image-segmentation","display_name":"Image segmentation","score":0.30300000309944153},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.29019999504089355}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5403000116348267},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5403000116348267},{"id":"https://openalex.org/C89600930","wikidata":"https://www.wikidata.org/wiki/Q1423946","display_name":"Segmentation","level":2,"score":0.46149998903274536},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.43140000104904175},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.35350000858306885},{"id":"https://openalex.org/C124504099","wikidata":"https://www.wikidata.org/wiki/Q56933","display_name":"Image segmentation","level":3,"score":0.30300000309944153},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.29019999504089355},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.274399995803833},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.26089999079704285},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.25130000710487366}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/iccp68926.2025.11427128","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccp68926.2025.11427128","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE 21st International Conference on Intelligent Computer Communication and Processing (ICCP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":7,"referenced_works":["https://openalex.org/W3121523901","https://openalex.org/W4312671791","https://openalex.org/W4391755841","https://openalex.org/W4401024413","https://openalex.org/W4403322060","https://openalex.org/W4405490600","https://openalex.org/W4410949532"],"related_works":[],"abstract_inverted_index":{"In":[0],"this":[1,59],"work,":[2],"we":[3,128],"propose":[4],"an":[5],"innovative":[6],"knowledge":[7,74],"distillation":[8,64,161],"framework":[9],"tailored":[10],"for":[11,37,142],"promptable":[12],"transformer-based":[13,28],"segmentation":[14,51,217],"models,":[15,25],"focusing":[16],"on":[17,87,147],"Segment":[18],"Anything":[19],"Model":[20],"2":[21],"(SAM":[22],"2).":[23],"Powerful":[24],"such":[26],"as":[27],"ones,":[29],"which":[30],"require":[31],"large-scale":[32],"datasets":[33],"and":[34,139,195],"extensive":[35],"computation":[36],"good":[38],"performance,":[39],"are":[40],"difficult":[41],"to":[42,71,145,162,193,197],"train,":[43],"even":[44,84],"with":[45,80,98,159,179,219],"limited":[46],"data,":[47],"often":[48],"yielding":[49],"uncertain":[50],"results":[52,209],"across":[53,204],"different":[54],"use":[55],"cases.":[56],"We":[57],"address":[58,117],"challenge":[60,119],"by":[61],"proposing":[62],"a":[63,68,76,130,184],"strategy":[65],"that":[66,134],"allows":[67],"student":[69,107,144,167,181],"model":[70,182],"inherit":[72],"structured":[73],"from":[75,95,101,125],"strong":[77],"pre-trained":[78],"teacher":[79,151],"the":[81,102,106,113,118,126,136,143,150,171,180,188],"same":[82],"architecture,":[83],"when":[85],"trained":[86],"small":[88],"datasets.":[89],"The":[90,155],"method":[91,175],"combines":[92],"hard":[93],"supervision":[94],"ground-truth":[96],"masks":[97],"soft":[99,137],"guidance":[100,141],"teacher\u2019s":[103],"logits,":[104],"helping":[105],"learn":[108],"meaningful":[109],"structures":[110],"early":[111],"in":[112,166,176,187,201],"training":[114],"process.":[115],"To":[116],"of":[120,173],"noisy":[121],"or":[122],"overconfident":[123],"predictions":[124],"teacher,":[127],"introduce":[129],"confidence-aware":[131],"agreement":[132],"mask":[133],"modulates":[135],"loss":[138,190],"provides":[140],"focus":[146],"regions":[148],"where":[149],"is":[152,157],"more":[153],"confident.":[154],"work":[156],"extended":[158],"token-level":[160],"enhance":[163],"structural":[164],"fidelity":[165],"representations.":[168],"Experiments":[169],"demonstrate":[170],"effectiveness":[172],"our":[174],"low-data":[177],"regimes,":[178],"achieving":[183],"notable":[185],"reduction":[186],"total":[189],"(from":[191],"15.17":[192],"8.84)":[194],"up":[196],"$20":[198],"\\%$":[199],"improvement":[200],"Intersection-over-Union":[202],"(IoU)":[203],"various":[205],"inference":[206],"scenarios.":[207],"Qualitative":[208],"further":[210],"support":[211],"these":[212],"findings,":[213],"showing":[214],"visibly":[215],"cleaner":[216],"outputs":[218],"reduced":[220],"noise,":[221],"indicating":[222],"improved":[223],"generalization":[224],"under":[225],"data-scarce":[226],"conditions.":[227]},"counts_by_year":[],"updated_date":"2026-03-15T07:15:06.534987","created_date":"2026-03-14T00:00:00"}