{"id":"https://openalex.org/W7161159406","doi":"https://doi.org/10.48550/arxiv.2605.12953","title":"Seg-Agent: Test-Time Multimodal Reasoning for Training-Free Language-Guided Segmentation","display_name":"Seg-Agent: Test-Time Multimodal Reasoning for Training-Free Language-Guided Segmentation","publication_year":2026,"publication_date":"2026-05-13","ids":{"openalex":"https://openalex.org/W7161159406","doi":"https://doi.org/10.48550/arxiv.2605.12953"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.12953","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.12953","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.2605.12953","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5021932420","display_name":"Chao Hao","orcid":"https://orcid.org/0000-0001-6700-9446"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hao, Chao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136127253","display_name":"Jun Xu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Jun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5132924900","display_name":"Ji Du","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Du, Ji","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5069824702","display_name":"Shuo Ye","orcid":"https://orcid.org/0000-0001-7756-8233"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ye, Shuo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136141745","display_name":"Ziyue Qiao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qiao, Ziyue","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136114907","display_name":"Xiaodong Cun","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cun, Xiaodong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136150635","display_name":"Guangcong Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Guangcong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136173384","display_name":"Xubin Zheng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zheng, Xubin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5136121522","display_name":"Zitong Yu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yu, Zitong","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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.986299991607666,"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":0.986299991607666,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.002199999988079071,"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":0.0008999999845400453,"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/visual-reasoning","display_name":"Visual reasoning","score":0.6759999990463257},{"id":"https://openalex.org/keywords/segmentation","display_name":"Segmentation","score":0.6432999968528748},{"id":"https://openalex.org/keywords/leverage","display_name":"Leverage (statistics)","score":0.6180999875068665},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.5094000101089478},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.4235999882221222},{"id":"https://openalex.org/keywords/natural-language","display_name":"Natural language","score":0.40070000290870667},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.3928999900817871},{"id":"https://openalex.org/keywords/language-understanding","display_name":"Language understanding","score":0.39250001311302185},{"id":"https://openalex.org/keywords/question-answering","display_name":"Question answering","score":0.38510000705718994}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7950999736785889},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6826000213623047},{"id":"https://openalex.org/C2777508537","wikidata":"https://www.wikidata.org/wiki/Q7936620","display_name":"Visual reasoning","level":2,"score":0.6759999990463257},{"id":"https://openalex.org/C89600930","wikidata":"https://www.wikidata.org/wiki/Q1423946","display_name":"Segmentation","level":2,"score":0.6432999968528748},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.6180999875068665},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.5094000101089478},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.4235999882221222},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4156000018119812},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.40070000290870667},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.3928999900817871},{"id":"https://openalex.org/C2983448237","wikidata":"https://www.wikidata.org/wiki/Q1078276","display_name":"Language understanding","level":2,"score":0.39250001311302185},{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.38510000705718994},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.375},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.3587000072002411},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.3375000059604645},{"id":"https://openalex.org/C2779439875","wikidata":"https://www.wikidata.org/wiki/Q1078276","display_name":"Natural language understanding","level":3,"score":0.3305000066757202},{"id":"https://openalex.org/C124504099","wikidata":"https://www.wikidata.org/wiki/Q56933","display_name":"Image segmentation","level":3,"score":0.3264000117778778},{"id":"https://openalex.org/C2780103172","wikidata":"https://www.wikidata.org/wiki/Q1309721","display_name":"Visual Objects","level":3,"score":0.3181000053882599},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3158999979496002},{"id":"https://openalex.org/C27511587","wikidata":"https://www.wikidata.org/wiki/Q2178623","display_name":"Spatial relation","level":2,"score":0.31029999256134033},{"id":"https://openalex.org/C2778012447","wikidata":"https://www.wikidata.org/wiki/Q1034415","display_name":"Scope (computer science)","level":2,"score":0.2957000136375427},{"id":"https://openalex.org/C155911833","wikidata":"https://www.wikidata.org/wiki/Q3817354","display_name":"Spatial intelligence","level":2,"score":0.29499998688697815},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.2766000032424927},{"id":"https://openalex.org/C2775955345","wikidata":"https://www.wikidata.org/wiki/Q7449071","display_name":"Semantic mapping","level":2,"score":0.27059999108314514},{"id":"https://openalex.org/C186644900","wikidata":"https://www.wikidata.org/wiki/Q194152","display_name":"Parsing","level":2,"score":0.2639999985694885},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.2572000026702881}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.12953","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.12953","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.2605.12953","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.12953","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":[{"score":0.7104874849319458,"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Language-guided":[0],"segmentation":[1,45,216],"transcends":[2],"the":[3,55,91,154,157,168,174,221],"scope":[4],"limitations":[5],"of":[6,60,225],"traditional":[7],"semantic":[8],"segmentation,":[9],"enabling":[10],"models":[11,46],"to":[12,35,49,54,73,84,148,159,183,187,196],"segment":[13],"arbitrary":[14],"target":[15],"regions":[16,151],"based":[17,98],"on":[18,67,70,99],"natural":[19],"language":[20],"instructions.":[21],"Existing":[22],"approaches":[23],"typically":[24],"adopt":[25],"a":[26,113,206],"two-stage":[27],"framework:":[28],"employing":[29],"Multimodal":[30,120],"Large":[31],"Language":[32],"Models":[33],"(MLLMs)":[34],"interpret":[36],"instructions":[37],"and":[38,139,161,214,223],"generate":[39],"visual":[40,105,131,146,169],"prompts,":[41],"followed":[42],"by":[43],"foundational":[44],"(e.g.,":[47],"SAM)":[48],"produce":[50],"masks.":[51],"However,":[52],"due":[53],"limited":[56],"spatial":[57,165],"grounding":[58],"capabilities":[59],"off-the-shelf":[61],"MLLMs,":[62],"these":[63],"methods":[64,190],"often":[65],"rely":[66],"extensive":[68],"training":[69],"large-scale":[71],"datasets":[72],"achieve":[74,184],"satisfactory":[75],"accuracy.":[76],"While":[77],"recent":[78],"advances":[79],"have":[80],"introduced":[81],"reasoning":[82,96,132],"mechanisms":[83],"improve":[85],"performance,":[86],"they":[87],"predominantly":[88],"operate":[89],"within":[90],"textual":[92,175],"domain,":[93],"performing":[94],"chain-of-thought":[95],"solely":[97],"abstract":[100],"text":[101],"representations":[102],"without":[103,191],"direct":[104],"feedback.":[106],"In":[107],"this":[108],"paper,":[109],"we":[110,142,203],"propose":[111],"Seg-Agent,":[112],"completely":[114],"training-free":[115],"framework":[116],"that":[117],"pioneers":[118],"Explicit":[119],"Chain-of-Reasoning.":[121],"Unlike":[122],"prior":[123],"text-only":[124],"reasoning,":[125],"our":[126,226],"approach":[127],"constructs":[128],"an":[129],"interactive":[130],"loop":[133],"comprising":[134],"three":[135],"stages:":[136],"generation,":[137],"selection,":[138],"refinement.":[140],"Specifically,":[141],"leverage":[143],"Set-of-Mark":[144],"(SoM)":[145],"prompting":[147],"render":[149],"candidate":[150],"directly":[152],"onto":[153],"image,":[155],"allowing":[156],"MLLM":[158],"``see''":[160],"iteratively":[162],"reason":[163],"about":[164],"relationships":[166],"in":[167],"domain":[170],"rather":[171],"than":[172],"just":[173],"one.":[176],"This":[177],"explicit":[178,210],"multimodal":[179],"interaction":[180],"enables":[181],"Seg-Agent":[182],"performance":[185],"comparable":[186],"state-of-the-art":[188],"training-based":[189],"any":[192],"parameter":[193],"updates.":[194],"Furthermore,":[195],"comprehensively":[197],"evaluate":[198],"generalization":[199],"across":[200],"diverse":[201],"scenarios,":[202],"introduce":[204],"Various-LangSeg,":[205],"novel":[207],"benchmark":[208],"covering":[209],"semantic,":[211],"generic":[212],"object,":[213],"reasoning-guided":[215],"tasks.":[217],"Extensive":[218],"experiments":[219],"demonstrate":[220],"effectiveness":[222],"robustness":[224],"method.":[227]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-15T00:00:00"}