{"id":"https://openalex.org/W7162686175","doi":"https://doi.org/10.48550/arxiv.2605.27959","title":"ROVER: Routing Object-Centric Visual Evidence for Grounded Multi-Image Reasoning","display_name":"ROVER: Routing Object-Centric Visual Evidence for Grounded Multi-Image Reasoning","publication_year":2026,"publication_date":"2026-05-27","ids":{"openalex":"https://openalex.org/W7162686175","doi":"https://doi.org/10.48550/arxiv.2605.27959"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.27959","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.27959","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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.2605.27959","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5127961887","display_name":"Guannan Lv","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lv, Guannan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137283745","display_name":"Ren Nie","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Nie, Ren","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137212748","display_name":"Hongjian Dou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dou, Hongjian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":null,"display_name":"Gao, Tingting","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gao, Tingting","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":0.9926000237464905,"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":0.9926000237464905,"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":0.0013000000035390258,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.0008999999845400453,"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/visual-reasoning","display_name":"Visual reasoning","score":0.7958999872207642},{"id":"https://openalex.org/keywords/object","display_name":"Object (grammar)","score":0.4724000096321106},{"id":"https://openalex.org/keywords/aggregate","display_name":"Aggregate (composite)","score":0.446399986743927},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.4341999888420105},{"id":"https://openalex.org/keywords/focus","display_name":"Focus (optics)","score":0.4172999858856201},{"id":"https://openalex.org/keywords/encoding","display_name":"Encoding (memory)","score":0.39169999957084656},{"id":"https://openalex.org/keywords/space","display_name":"Space (punctuation)","score":0.3896999955177307},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.375},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.3598000109195709},{"id":"https://openalex.org/keywords/selection","display_name":"Selection (genetic algorithm)","score":0.3513999879360199}],"concepts":[{"id":"https://openalex.org/C2777508537","wikidata":"https://www.wikidata.org/wiki/Q7936620","display_name":"Visual reasoning","level":2,"score":0.7958999872207642},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7445999979972839},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.48910000920295715},{"id":"https://openalex.org/C2781238097","wikidata":"https://www.wikidata.org/wiki/Q175026","display_name":"Object (grammar)","level":2,"score":0.4724000096321106},{"id":"https://openalex.org/C4679612","wikidata":"https://www.wikidata.org/wiki/Q866298","display_name":"Aggregate (composite)","level":2,"score":0.446399986743927},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.4341999888420105},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.4172999858856201},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.39169999957084656},{"id":"https://openalex.org/C2778572836","wikidata":"https://www.wikidata.org/wiki/Q380933","display_name":"Space (punctuation)","level":2,"score":0.3896999955177307},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.375},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.3598000109195709},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.3513999879360199},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.350600004196167},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.3504999876022339},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.33730000257492065},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.33379998803138733},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.3327000141143799},{"id":"https://openalex.org/C178253425","wikidata":"https://www.wikidata.org/wiki/Q162668","display_name":"Visual perception","level":3,"score":0.2953999936580658},{"id":"https://openalex.org/C4924752","wikidata":"https://www.wikidata.org/wiki/Q184148","display_name":"Plug-in","level":2,"score":0.29319998621940613},{"id":"https://openalex.org/C64876066","wikidata":"https://www.wikidata.org/wiki/Q5141226","display_name":"Cognitive neuroscience of visual object recognition","level":3,"score":0.2906000018119812},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.29019999504089355},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.2897000014781952},{"id":"https://openalex.org/C207363949","wikidata":"https://www.wikidata.org/wiki/Q462915","display_name":"Visual space","level":3,"score":0.28619998693466187},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.2847999930381775},{"id":"https://openalex.org/C158495155","wikidata":"https://www.wikidata.org/wiki/Q2369151","display_name":"Visual search","level":2,"score":0.2800000011920929},{"id":"https://openalex.org/C2776505523","wikidata":"https://www.wikidata.org/wiki/Q4785468","display_name":"Plan (archaeology)","level":2,"score":0.2766000032424927},{"id":"https://openalex.org/C93226319","wikidata":"https://www.wikidata.org/wiki/Q193137","display_name":"Differential (mechanical device)","level":2,"score":0.27309998869895935},{"id":"https://openalex.org/C13606891","wikidata":"https://www.wikidata.org/wiki/Q2623243","display_name":"Conceptual model","level":2,"score":0.2694999873638153},{"id":"https://openalex.org/C74172769","wikidata":"https://www.wikidata.org/wiki/Q1446839","display_name":"Routing (electronic design automation)","level":2,"score":0.2685000002384186},{"id":"https://openalex.org/C42058472","wikidata":"https://www.wikidata.org/wiki/Q810214","display_name":"Base (topology)","level":2,"score":0.2612000107765198},{"id":"https://openalex.org/C56461940","wikidata":"https://www.wikidata.org/wiki/Q970687","display_name":"Eye tracking","level":2,"score":0.2563999891281128},{"id":"https://openalex.org/C83665646","wikidata":"https://www.wikidata.org/wiki/Q42139305","display_name":"Feature vector","level":2,"score":0.2533999979496002},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.2524000108242035}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.27959","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.27959","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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.2605.27959","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.27959","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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":{"Multimodal":[0],"Large":[1],"Language":[2],"Models":[3],"(MLLMs)":[4],"have":[5],"increasingly":[6],"localized":[7],"and":[8,44,56,129,132,138,151,164,181],"interleaved":[9,154],"visual":[10,62,94,122],"evidence":[11,95,135],"for":[12,83,91,143],"deliberative":[13],"reasoning.":[14,145],"Grounding-based":[15],"approaches":[16],"typically":[17],"focus":[18],"on":[19,173,198],"regions":[20],"of":[21,58],"interest":[22],"(RoIs)":[23],"by":[24,196],"injecting":[25],"cropped":[26],"image":[27],"patches":[28],"or":[29,69],"RoI-specific":[30],"features":[31],"into":[32,120,149],"the":[33,54,112,161,170,193],"reasoning":[34,114],"context.":[35],"However,":[36],"such":[37],"designs":[38],"can":[39],"weaken":[40],"holistic":[41],"scene":[42],"understanding":[43],"inter-object":[45],"relations,":[46],"while":[47],"incurring":[48],"decoding":[49],"costs":[50],"that":[51],"scale":[52],"with":[53],"number":[55],"size":[57],"RoIs.":[59],"Alternatively,":[60],"adaptive":[61],"feature":[63],"selection":[64],"often":[65],"requires":[66],"fine-grained":[67],"supervision":[68],"complex":[70],"heuristics.":[71],"To":[72],"address":[73],"these":[74],"limitations,":[75],"we":[76],"propose":[77],"ROVER":[78,102,148],"(Routing":[79],"Object-centric":[80],"Visual":[81],"Evidence":[82],"grounded":[84],"multi-image":[85],"Reasoning),":[86],"a":[87,104,121],"lightweight,":[88],"learnable":[89],"plugin":[90],"efficient":[92],"global":[93],"routing.":[96],"Upon":[97],"each":[98],"object":[99],"grounding":[100,179],"prediction,":[101],"injects":[103],"step-specific":[105],"token":[106],"triplet":[107],"to":[108,160],"synergistically:":[109],"(i)":[110],"aggregate":[111],"ongoing":[113],"context,":[115],"(ii)":[116],"distill":[117],"intra-image":[118],"cues":[119],"working":[123],"space":[124,142],"via":[125],"object-centric":[126],"differential":[127],"attention,":[128],"(iii)":[130],"route":[131],"integrate":[133,147],"history-aware":[134],"across":[136,200],"objects":[137],"images":[139],"within":[140],"this":[141],"subsequent":[144],"We":[146],"Qwen2.5-VL-7B":[150],"develop":[152],"an":[153],"SFT-to-GRPO":[155],"training":[156],"pipeline.":[157],"Strictly":[158],"adhering":[159],"original":[162],"datasets":[163],"evaluation":[165],"protocols,":[166],"our":[167],"method":[168],"achieves":[169],"best":[171],"performance":[172],"MM-GCoT":[174],"(+4.8%":[175],"answer":[176,184],"accuracy,":[177],"+14.6%":[178],"accuracy)":[180],"VideoEspresso":[182],"(+8.6%":[183],"accuracy).":[185],"The":[186],"VideoEspresso-trained":[187],"model":[188,195],"demonstrates":[189],"strong":[190],"transferability,":[191],"outperforming":[192],"base":[194],"+4.7%":[197],"average":[199],"diverse":[201],"benchmarks.":[202]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-05-29T00:00:00"}