{"id":"https://openalex.org/W7151542845","doi":"https://doi.org/10.48550/arxiv.2604.03307","title":"V-Reflection: Transforming MLLMs from Passive Observers to Active Interrogators","display_name":"V-Reflection: Transforming MLLMs from Passive Observers to Active Interrogators","publication_year":2026,"publication_date":"2026-03-31","ids":{"openalex":"https://openalex.org/W7151542845","doi":"https://doi.org/10.48550/arxiv.2604.03307"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.03307","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.03307","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.2604.03307","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5074096594","display_name":"Jiazhou Zhou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Jiazhou","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128112520","display_name":"Yucheng Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Yucheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133102028","display_name":"Hongyang Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Hongyang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133112770","display_name":"Qing Jiang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiang, Qing","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133079970","display_name":"Hu Zhou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Hu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133073730","display_name":"Ying-Cong Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Ying-Cong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5133108197","display_name":"Lei Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Lei","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":0.30239999294281006,"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":0.30239999294281006,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.258899986743927,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.045499999076128006,"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/feature","display_name":"Feature (linguistics)","score":0.5493999719619751},{"id":"https://openalex.org/keywords/perception","display_name":"Perception","score":0.5213000178337097},{"id":"https://openalex.org/keywords/visual-space","display_name":"Visual space","score":0.49869999289512634},{"id":"https://openalex.org/keywords/function","display_name":"Function (biology)","score":0.477400004863739},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.4717000126838684},{"id":"https://openalex.org/keywords/decoding-methods","display_name":"Decoding methods","score":0.4690999984741211},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.4277999997138977},{"id":"https://openalex.org/keywords/visualization","display_name":"Visualization","score":0.42340001463890076},{"id":"https://openalex.org/keywords/encoding","display_name":"Encoding (memory)","score":0.4189000129699707},{"id":"https://openalex.org/keywords/autoregressive-model","display_name":"Autoregressive model","score":0.39070001244544983}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6502000093460083},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.5493999719619751},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5249999761581421},{"id":"https://openalex.org/C26760741","wikidata":"https://www.wikidata.org/wiki/Q160402","display_name":"Perception","level":2,"score":0.5213000178337097},{"id":"https://openalex.org/C207363949","wikidata":"https://www.wikidata.org/wiki/Q462915","display_name":"Visual space","level":3,"score":0.49869999289512634},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.477400004863739},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.4717000126838684},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.4690999984741211},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.4449000060558319},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.4277999997138977},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.42340001463890076},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.4189000129699707},{"id":"https://openalex.org/C159877910","wikidata":"https://www.wikidata.org/wiki/Q2202883","display_name":"Autoregressive model","level":2,"score":0.39070001244544983},{"id":"https://openalex.org/C178253425","wikidata":"https://www.wikidata.org/wiki/Q162668","display_name":"Visual perception","level":3,"score":0.3905999958515167},{"id":"https://openalex.org/C2777508537","wikidata":"https://www.wikidata.org/wiki/Q7936620","display_name":"Visual reasoning","level":2,"score":0.38940000534057617},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.3702000081539154},{"id":"https://openalex.org/C2776010242","wikidata":"https://www.wikidata.org/wiki/Q4677575","display_name":"Active perception","level":3,"score":0.35929998755455017},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.3483999967575073},{"id":"https://openalex.org/C65682993","wikidata":"https://www.wikidata.org/wiki/Q1056451","display_name":"Reflection (computer programming)","level":2,"score":0.3375000059604645},{"id":"https://openalex.org/C131970427","wikidata":"https://www.wikidata.org/wiki/Q17028356","display_name":"Hyperacuity","level":2,"score":0.3366999924182892},{"id":"https://openalex.org/C49937458","wikidata":"https://www.wikidata.org/wiki/Q2599292","display_name":"Probabilistic logic","level":2,"score":0.3319000005722046},{"id":"https://openalex.org/C88516994","wikidata":"https://www.wikidata.org/wiki/Q1268863","display_name":"Dynamic time warping","level":2,"score":0.33059999346733093},{"id":"https://openalex.org/C2778833722","wikidata":"https://www.wikidata.org/wiki/Q28457372","display_name":"Tact","level":2,"score":0.3140000104904175},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.31029999256134033},{"id":"https://openalex.org/C83665646","wikidata":"https://www.wikidata.org/wiki/Q42139305","display_name":"Feature vector","level":2,"score":0.3037000000476837},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.29409998655319214},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.28029999136924744},{"id":"https://openalex.org/C124304363","wikidata":"https://www.wikidata.org/wiki/Q673661","display_name":"Abstraction","level":2,"score":0.2791999876499176},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.27869999408721924},{"id":"https://openalex.org/C134400042","wikidata":"https://www.wikidata.org/wiki/Q2372244","display_name":"Symbol (formal)","level":2,"score":0.2768000066280365},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.27239999175071716},{"id":"https://openalex.org/C2780691134","wikidata":"https://www.wikidata.org/wiki/Q1129448","display_name":"Preamble","level":3,"score":0.26989999413490295},{"id":"https://openalex.org/C2778572836","wikidata":"https://www.wikidata.org/wiki/Q380933","display_name":"Space (punctuation)","level":2,"score":0.2678000032901764},{"id":"https://openalex.org/C180016635","wikidata":"https://www.wikidata.org/wiki/Q2712821","display_name":"Compression (physics)","level":2,"score":0.2581999897956848},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.25679999589920044},{"id":"https://openalex.org/C160086991","wikidata":"https://www.wikidata.org/wiki/Q5939193","display_name":"Human visual system model","level":3,"score":0.2567000091075897}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.03307","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.03307","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.2604.03307","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.03307","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":{"Multimodal":[0],"Large":[1],"Language":[2],"Models":[3],"(MLLMs)":[4],"have":[5],"achieved":[6],"remarkable":[7],"success,":[8],"yet":[9],"they":[10],"remain":[11,178],"prone":[12],"to":[13,31,56,60,170],"perception-related":[14],"hallucinations":[15],"in":[16,187],"fine-grained":[17,209],"tasks.":[18],"This":[19],"vulnerability":[20],"arises":[21],"from":[22],"a":[23,39,45,72,83,113,132,182],"fundamental":[24],"limitation:":[25],"their":[26,62],"reasoning":[27,64,105,216],"is":[28],"largely":[29],"restricted":[30],"the":[32,76,99,118,139,148,155,159,163,168,188,197,208],"language":[33],"domain,":[34],"treating":[35],"visual":[36,58,85,100,150,220],"input":[37],"as":[38,52,93],"static,":[40],"reasoning-agnostic":[41],"preamble":[42],"rather":[43],"than":[44],"dynamic":[46,94,144],"participant.":[47],"Consequently,":[48],"current":[49],"models":[50],"act":[51],"passive":[53],"observers,":[54],"unable":[55],"re-examine":[57],"details":[59],"ground":[61],"evolving":[63],"states.":[65],"To":[66],"overcome":[67],"this,":[68],"we":[69],"propose":[70],"V-Reflection,":[71],"framework":[73],"that":[74,96,146,214],"transforms":[75],"MLLM":[77],"into":[78,143,162],"an":[79],"active":[80],"interrogator":[81],"through":[82,127],"\"think-then-look\"":[84],"reflection":[86],"mechanism.":[87],"During":[88,174],"reasoning,":[89],"latent":[90,189,215],"states":[91,142],"function":[92],"probes":[95,145],"actively":[97],"interrogate":[98,147],"feature":[101,151],"space,":[102],"grounding":[103],"each":[104],"step":[106],"for":[107],"task-critical":[108,172,219],"evidence.":[109,173,221],"Our":[110],"approach":[111],"employs":[112],"two-stage":[114],"distillation":[115],"strategy.":[116],"First,":[117],"Box-Guided":[119],"Compression":[120,135],"Module":[121],"(BCM)":[122],"establishes":[123],"stable":[124],"pixel-to-latent":[125],"targets":[126],"explicit":[128],"spatial":[129,156],"grounding.":[130],"Next,":[131],"Dynamic":[133],"Autoregressive":[134],"(DAC)":[136],"module":[137],"maps":[138],"model's":[140],"hidden":[141],"global":[149],"map.":[152],"By":[153],"distilling":[154],"expertise":[157],"of":[158,199],"BCM":[160],"teacher":[161],"DAC":[164],"student,":[165],"V-Reflection":[166,201],"internalizes":[167],"ability":[169],"localize":[171],"inference,":[175],"both":[176],"modules":[177],"entirely":[179],"inactive,":[180],"maintaining":[181],"purely":[183],"end-to-end":[184],"autoregressive":[185],"decoding":[186],"space":[190],"with":[191],"optimal":[192],"efficiency.":[193],"Extensive":[194],"experiments":[195],"demonstrate":[196],"effectiveness":[198],"our":[200],"across":[202],"six":[203],"perception-intensive":[204],"benchmarks,":[205],"significantly":[206],"narrowing":[207],"perception":[210],"gap.":[211],"Visualizations":[212],"confirm":[213],"autonomously":[217],"localizes":[218]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-04-08T00:00:00"}