{"id":"https://openalex.org/W7160944343","doi":"https://doi.org/10.48550/arxiv.2605.08802","title":"CoLVR: Enhancing Exploratory Latent Visual Reasoning via Contrastive Optimization","display_name":"CoLVR: Enhancing Exploratory Latent Visual Reasoning via Contrastive Optimization","publication_year":2026,"publication_date":"2026-05-09","ids":{"openalex":"https://openalex.org/W7160944343","doi":"https://doi.org/10.48550/arxiv.2605.08802"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.08802","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.08802","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":null,"license_id":null,"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.2605.08802","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5135938074","display_name":"Ziyang Ding","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ding, Ziyang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5049175214","display_name":"Linjian Meng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Meng, Linjian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135997470","display_name":"Yiming Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Yiming","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135926499","display_name":"Yuhan Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Yuhan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135941728","display_name":"Yuhao Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Yuhao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5135918137","display_name":"Zhen Zhao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhao, Zhen","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":0.9858999848365784,"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":0.9858999848365784,"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/T11273","display_name":"Advanced Graph Neural Networks","score":0.0017000000225380063,"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":0.00139999995008111,"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/latent-variable","display_name":"Latent variable","score":0.6233999729156494},{"id":"https://openalex.org/keywords/visual-reasoning","display_name":"Visual reasoning","score":0.5697000026702881},{"id":"https://openalex.org/keywords/latent-semantic-analysis","display_name":"Latent semantic analysis","score":0.5180000066757202},{"id":"https://openalex.org/keywords/probabilistic-latent-semantic-analysis","display_name":"Probabilistic latent semantic analysis","score":0.4875999987125397},{"id":"https://openalex.org/keywords/latent-variable-model","display_name":"Latent variable model","score":0.4352000057697296},{"id":"https://openalex.org/keywords/domain","display_name":"Domain (mathematical analysis)","score":0.41339999437332153},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.3944000005722046}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6848999857902527},{"id":"https://openalex.org/C51167844","wikidata":"https://www.wikidata.org/wiki/Q4422623","display_name":"Latent variable","level":2,"score":0.6233999729156494},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6086000204086304},{"id":"https://openalex.org/C2777508537","wikidata":"https://www.wikidata.org/wiki/Q7936620","display_name":"Visual reasoning","level":2,"score":0.5697000026702881},{"id":"https://openalex.org/C170133592","wikidata":"https://www.wikidata.org/wiki/Q1806883","display_name":"Latent semantic analysis","level":2,"score":0.5180000066757202},{"id":"https://openalex.org/C112933361","wikidata":"https://www.wikidata.org/wiki/Q2845258","display_name":"Probabilistic latent semantic analysis","level":2,"score":0.4875999987125397},{"id":"https://openalex.org/C65965080","wikidata":"https://www.wikidata.org/wiki/Q1806885","display_name":"Latent variable model","level":3,"score":0.4352000057697296},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.41339999437332153},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.3944000005722046},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.38519999384880066},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.3707999885082245},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3513000011444092},{"id":"https://openalex.org/C188198153","wikidata":"https://www.wikidata.org/wiki/Q1613840","display_name":"Limiting","level":2,"score":0.3418999910354614},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.2727999985218048},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.26989999413490295},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.26969999074935913},{"id":"https://openalex.org/C70727504","wikidata":"https://www.wikidata.org/wiki/Q1806878","display_name":"Latent class model","level":2,"score":0.2630999982357025},{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.26249998807907104}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.08802","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.08802","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.08802","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.08802","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":null,"license_id":null,"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":{"Due":[0],"to":[1,14,21,47,51,131],"the":[2,59,110,152],"potential":[3],"for":[4,74,126],"exploratory":[5,60,82,97,153],"reasoning":[6,24,63,138,144],"of":[7,31,61,135,155,161,177],"Latent":[8,75],"Visual":[9,76],"Reasoning,":[10],"recent":[11],"works":[12,40],"tend":[13],"enable":[15,132],"MLLMs":[16],"(Multimodal":[17],"Large":[18],"Language":[19],"Models)":[20],"perform":[22],"visual":[23,54,83,137],"by":[25,105],"propagating":[26],"continuous":[27],"hidden":[28],"states":[29],"instead":[30],"decoding":[32],"intermediate":[33],"steps":[34],"into":[35],"discrete":[36],"tokens.":[37],"However,":[38],"existing":[39,172],"typically":[41],"rely":[42],"on":[43,163,167,175,184],"hard":[44],"alignment":[45],"objectives":[46],"force":[48],"latent":[49,62,88,101,112,122,136,156,173],"representations":[50,98],"match":[52],"predefined":[53],"features,":[55],"thereby":[56],"severely":[57],"limiting":[58],"process.":[64],"To":[65,78],"address":[66],"this":[67],"problem,":[68],"we":[69],"propose":[70],"CoLVR":[71,85,93,119,149],"(Contrastive":[72],"Optimization":[73],"Reasoning).":[77],"obtain":[79],"a":[80,87,100,121,181],"more":[81],"reasoning,":[84],"introduces":[86],"contrastive":[89,102,124],"training":[90],"framework.":[91],"Firstly,":[92],"learns":[94],"diverse":[95,143],"and":[96,114,140,165,189],"with":[99,180],"objective":[103],"guided":[104],"angle-based":[106],"perturbation,":[107],"which":[108],"expands":[109],"semantic":[111],"space":[113],"avoids":[115],"over-constrained":[116],"embedding.":[117],"Then,":[118],"employs":[120],"trajectory":[123],"reward":[125],"RL":[127],"(Reinforcement":[128],"Learning)":[129],"post-training":[130],"fine-grained":[133],"optimization":[134],"process":[139],"thus":[141],"fostering":[142],"behaviors.":[145],"Experiments":[146],"demonstrate":[147],"that":[148],"significantly":[150],"enhances":[151],"capability":[154],"representations,":[157],"achieving":[158],"average":[159],"improvements":[160],"5.83%":[162],"VSP":[164],"8.00%":[166],"Jigsaw,":[168],"while":[169],"also":[170],"outperforming":[171],"models":[174,190],"out":[176],"domain":[178],"benchmarks,":[179],"3.40%":[182],"gain":[183],"MMStar.":[185],"The":[186],"data,":[187],"codes,":[188],"are":[191],"released":[192],"at":[193],"https://github.com/Oscar-dzy/CoLVR.":[194]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-13T00:00:00"}