{"id":"https://openalex.org/W7140303621","doi":"https://doi.org/10.48550/arxiv.2603.23455","title":"DetPO: In-Context Learning with Multi-Modal LLMs for Few-Shot Object Detection","display_name":"DetPO: In-Context Learning with Multi-Modal LLMs for Few-Shot Object Detection","publication_year":2026,"publication_date":"2026-03-24","ids":{"openalex":"https://openalex.org/W7140303621","doi":"https://doi.org/10.48550/arxiv.2603.23455"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.23455","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.23455","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":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2603.23455","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5022291473","display_name":"Gautam Rajendrakumar Gare","orcid":"https://orcid.org/0000-0002-1689-9626"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gare, Gautam Rajendrakumar","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5006107116","display_name":"Neehar Peri","orcid":"https://orcid.org/0000-0001-9329-3989"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Peri, Neehar","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5104300701","display_name":"Matvei Popov","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Popov, Matvei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130608322","display_name":"Shruti Jain","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jain, Shruti","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5051265325","display_name":"John Galeotti","orcid":"https://orcid.org/0000-0003-4247-4311"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Galeotti, John","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5130560596","display_name":"Deva Ramanan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ramanan, Deva","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.5275999903678894,"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"}},"topics":[{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.5275999903678894,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.30379998683929443,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.06920000165700912,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/leverage","display_name":"Leverage (statistics)","score":0.7608000040054321},{"id":"https://openalex.org/keywords/object-detection","display_name":"Object detection","score":0.7515000104904175},{"id":"https://openalex.org/keywords/object","display_name":"Object (grammar)","score":0.5440000295639038},{"id":"https://openalex.org/keywords/class","display_name":"Class (philosophy)","score":0.5033000111579895},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.40959998965263367},{"id":"https://openalex.org/keywords/modalities","display_name":"Modalities","score":0.4075999855995178},{"id":"https://openalex.org/keywords/visual-reasoning","display_name":"Visual reasoning","score":0.34610000252723694}],"concepts":[{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.7608000040054321},{"id":"https://openalex.org/C2776151529","wikidata":"https://www.wikidata.org/wiki/Q3045304","display_name":"Object detection","level":3,"score":0.7515000104904175},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7491999864578247},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.7031000256538391},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5888000130653381},{"id":"https://openalex.org/C2781238097","wikidata":"https://www.wikidata.org/wiki/Q175026","display_name":"Object (grammar)","level":2,"score":0.5440000295639038},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.5033000111579895},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.41429999470710754},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.40959998965263367},{"id":"https://openalex.org/C2779903281","wikidata":"https://www.wikidata.org/wiki/Q6888026","display_name":"Modalities","level":2,"score":0.4075999855995178},{"id":"https://openalex.org/C2777508537","wikidata":"https://www.wikidata.org/wiki/Q7936620","display_name":"Visual reasoning","level":2,"score":0.34610000252723694},{"id":"https://openalex.org/C64876066","wikidata":"https://www.wikidata.org/wiki/Q5141226","display_name":"Cognitive neuroscience of visual object recognition","level":3,"score":0.33410000801086426},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.3199999928474426},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.3066999912261963},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.28209999203681946},{"id":"https://openalex.org/C2776036281","wikidata":"https://www.wikidata.org/wiki/Q48769818","display_name":"Constraint (computer-aided design)","level":2,"score":0.28029999136924744},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.2721000015735626},{"id":"https://openalex.org/C71681937","wikidata":"https://www.wikidata.org/wiki/Q3045304","display_name":"Object-class detection","level":5,"score":0.26910001039505005},{"id":"https://openalex.org/C12267149","wikidata":"https://www.wikidata.org/wiki/Q282453","display_name":"Support vector machine","level":2,"score":0.2587999999523163}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.23455","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.23455","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":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2603.23455","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.23455","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":false,"raw_source_name":null,"raw_type":"Preprint"},"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":{"Multi-Modal":[0],"LLMs":[1],"(MLLMs)":[2],"demonstrate":[3],"strong":[4],"visual":[5,75,139],"grounding":[6],"capabilities":[7],"on":[8,102,137,155],"popular":[9],"object":[10,82,113],"detection":[11,57,135],"benchmarks":[12],"like":[13],"OdinW-13":[14],"and":[15,28,77,93,157,170],"RefCOCO.":[16],"However,":[17],"state-of-the-art":[18,94],"models":[19,96],"still":[20],"struggle":[21],"to":[22,24,44,100,165],"generalize":[23],"out-of-distribution":[25],"classes,":[26],"tasks":[27],"imaging":[29],"modalities":[30],"not":[31],"typically":[32,88],"found":[33],"in":[34],"their":[35],"pre-training.":[36],"While":[37],"in-context":[38],"prompting":[39,60],"is":[40],"a":[41,124],"common":[42],"strategy":[43],"improve":[45],"performance":[46],"across":[47,152],"diverse":[48],"tasks,":[49],"we":[50,105,118],"find":[51],"that":[52,67,129],"it":[53],"often":[54],"yields":[55,149],"lower":[56],"accuracy":[58,136],"than":[59],"with":[61],"class":[62],"names":[63],"alone.":[64],"This":[65],"suggests":[66],"current":[68],"MLLMs":[69,86,154],"cannot":[70],"yet":[71],"effectively":[72],"leverage":[73],"few-shot":[74,112,138],"examples":[76,141],"rich":[78],"textual":[79],"descriptions":[80],"for":[81,111],"detection.":[83,114],"Since":[84],"frontier":[85],"are":[87,97,173],"only":[89],"accessible":[90],"via":[91],"APIs,":[92],"open-weights":[95],"prohibitively":[98],"expensive":[99],"fine-tune":[101],"consumer-grade":[103],"hardware,":[104],"instead":[106],"explore":[107],"black-box":[108,161],"prompt":[109],"optimization":[110,127],"To":[115],"this":[116],"end,":[117],"propose":[119],"Detection":[120],"Prompt":[121],"Optimization":[122],"(DetPO),":[123],"gradient-free":[125],"test-time":[126],"approach":[128,148],"refines":[130],"text-only":[131],"prompts":[132,172],"by":[133,163],"maximizing":[134],"training":[140],"while":[142],"calibrating":[143],"prediction":[144],"confidence.":[145],"Our":[146,168],"proposed":[147],"consistent":[150],"improvements":[151],"generalist":[153],"Roboflow20-VL":[156],"LVIS,":[158],"outperforming":[159],"prior":[160],"approaches":[162],"up":[164],"9.7":[166],"mAP.":[167],"code":[169],"optimized":[171],"available":[174],"at":[175],"https://ggare-cmu.github.io/DetPO/":[176]},"counts_by_year":[],"updated_date":"2026-07-02T06:12:58.138171","created_date":"2026-03-26T00:00:00"}