{"id":"https://openalex.org/W7123568614","doi":"https://doi.org/10.48550/arxiv.2601.06803","title":"Forest Before Trees: Latent Superposition for Efficient Visual Reasoning","display_name":"Forest Before Trees: Latent Superposition for Efficient Visual Reasoning","publication_year":2026,"publication_date":"2026-01-11","ids":{"openalex":"https://openalex.org/W7123568614","doi":"https://doi.org/10.48550/arxiv.2601.06803"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2601.06803","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2601.06803","pdf_url":"https://arxiv.org/pdf/2601.06803","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,"listed_in":[],"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-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2601.06803","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5102759447","display_name":"Yubo Wang","orcid":"https://orcid.org/0009-0004-6288-0880","observed_orcids":["https://orcid.org/0009-0004-6288-0880"]},"institutions":[{"id":"https://openalex.org/I136199984","display_name":"Harvard University","ror":"https://ror.org/03vek6s52","country_code":"US","type":"education","lineage":["https://openalex.org/I136199984"]},{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]},{"id":"https://openalex.org/I78988378","display_name":"Renmin University of China","ror":"https://ror.org/041pakw92","country_code":"CN","type":"education","lineage":["https://openalex.org/I78988378"]}],"countries":["CN","US"],"is_corresponding":false,"raw_author_name":"Wang, Yubo","raw_affiliation_strings":["Fudan University \u2663 Gaoling School of Artificial Intelligence , Renmin University of China \u2662 Harvard University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fudan University \u2663 Gaoling School of Artificial Intelligence , Renmin University of China \u2662 Harvard University","institution_ids":["https://openalex.org/I136199984","https://openalex.org/I24943067","https://openalex.org/I78988378"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5153460913","display_name":"Juntian Zhang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I136199984","display_name":"Harvard University","ror":"https://ror.org/03vek6s52","country_code":"US","type":"education","lineage":["https://openalex.org/I136199984"]},{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]},{"id":"https://openalex.org/I78988378","display_name":"Renmin University of China","ror":"https://ror.org/041pakw92","country_code":"CN","type":"education","lineage":["https://openalex.org/I78988378"]}],"countries":["CN","US"],"is_corresponding":false,"raw_author_name":"Zhang, Juntian","raw_affiliation_strings":["Fudan University \u2663 Gaoling School of Artificial Intelligence , Renmin University of China \u2662 Harvard University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fudan University \u2663 Gaoling School of Artificial Intelligence , Renmin University of China \u2662 Harvard University","institution_ids":["https://openalex.org/I136199984","https://openalex.org/I24943067","https://openalex.org/I78988378"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5148877141","display_name":"Yichen Wu","orcid":"https://orcid.org/0000-0003-2859-3285","observed_orcids":["https://orcid.org/0000-0003-2859-3285"]},"institutions":[{"id":"https://openalex.org/I136199984","display_name":"Harvard University","ror":"https://ror.org/03vek6s52","country_code":"US","type":"education","lineage":["https://openalex.org/I136199984"]},{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]},{"id":"https://openalex.org/I78988378","display_name":"Renmin University of China","ror":"https://ror.org/041pakw92","country_code":"CN","type":"education","lineage":["https://openalex.org/I78988378"]}],"countries":["CN","US"],"is_corresponding":false,"raw_author_name":"Wu, Yichen","raw_affiliation_strings":["Fudan University \u2663 Gaoling School of Artificial Intelligence , Renmin University of China \u2662 Harvard University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fudan University \u2663 Gaoling School of Artificial Intelligence , Renmin University of China \u2662 Harvard University","institution_ids":["https://openalex.org/I136199984","https://openalex.org/I24943067","https://openalex.org/I78988378"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125430101","display_name":"Yankai Lin","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I136199984","display_name":"Harvard University","ror":"https://ror.org/03vek6s52","country_code":"US","type":"education","lineage":["https://openalex.org/I136199984"]},{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]},{"id":"https://openalex.org/I78988378","display_name":"Renmin University of China","ror":"https://ror.org/041pakw92","country_code":"CN","type":"education","lineage":["https://openalex.org/I78988378"]}],"countries":["CN","US"],"is_corresponding":false,"raw_author_name":"Lin, Yankai","raw_affiliation_strings":["Fudan University \u2663 Gaoling School of Artificial Intelligence , Renmin University of China \u2662 Harvard University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fudan University \u2663 Gaoling School of Artificial Intelligence , Renmin University of China \u2662 Harvard University","institution_ids":["https://openalex.org/I136199984","https://openalex.org/I24943067","https://openalex.org/I78988378"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5086633938","display_name":"Nils Lukas","orcid":"https://orcid.org/0009-0001-5891-9154","observed_orcids":["https://orcid.org/0009-0001-5891-9154"]},"institutions":[{"id":"https://openalex.org/I136199984","display_name":"Harvard University","ror":"https://ror.org/03vek6s52","country_code":"US","type":"education","lineage":["https://openalex.org/I136199984"]},{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]},{"id":"https://openalex.org/I78988378","display_name":"Renmin University of China","ror":"https://ror.org/041pakw92","country_code":"CN","type":"education","lineage":["https://openalex.org/I78988378"]}],"countries":["CN","US"],"is_corresponding":false,"raw_author_name":"Lukas, Nils","raw_affiliation_strings":["Fudan University \u2663 Gaoling School of Artificial Intelligence , Renmin University of China \u2662 Harvard University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fudan University \u2663 Gaoling School of Artificial Intelligence , Renmin University of China \u2662 Harvard University","institution_ids":["https://openalex.org/I136199984","https://openalex.org/I24943067","https://openalex.org/I78988378"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5122962308","display_name":"Yuhan Liu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I136199984","display_name":"Harvard University","ror":"https://ror.org/03vek6s52","country_code":"US","type":"education","lineage":["https://openalex.org/I136199984"]},{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]},{"id":"https://openalex.org/I78988378","display_name":"Renmin University of China","ror":"https://ror.org/041pakw92","country_code":"CN","type":"education","lineage":["https://openalex.org/I78988378"]}],"countries":["CN","US"],"is_corresponding":false,"raw_author_name":"Liu, Yuhan","raw_affiliation_strings":["Fudan University \u2663 Gaoling School of Artificial Intelligence , Renmin University of China \u2662 Harvard University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fudan University \u2663 Gaoling School of Artificial Intelligence , Renmin University of China \u2662 Harvard University","institution_ids":["https://openalex.org/I136199984","https://openalex.org/I24943067","https://openalex.org/I78988378"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"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":1.0,"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":1.0,"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":0.0,"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/T11605","display_name":"Visual Attention and Saliency Detection","score":0.0,"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-reasoning","display_name":"latent reasoning","score":0.675000011920929},{"id":"https://openalex.org/keywords/visual-reasoning","display_name":"visual reasoning","score":0.7590000033378601},{"id":"https://openalex.org/keywords/large-vision-language-models","display_name":"large vision-language models","score":0.8410000205039978},{"id":"https://openalex.org/keywords/chain-of-thought","display_name":"chain-of-thought","score":0.843999981880188},{"id":"https://openalex.org/keywords/semantic-collapse","display_name":"semantic collapse","score":0.5490000247955322},{"id":"https://openalex.org/keywords/interpretability","display_name":"interpretability","score":0.7020000219345093},{"id":"https://openalex.org/keywords/inference-efficiency","display_name":"inference efficiency","score":0.7049999833106995},{"id":"https://openalex.org/keywords/out-of-distribution-generalization","display_name":"out-of-distribution generalization","score":0.8330000042915344}],"concepts":[{"id":"https://openalex.org/C2781067378","wikidata":"https://www.wikidata.org/wiki/Q17027399","display_name":"Interpretability","level":2,"score":0.7583000063896179},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6639000177383423},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6033999919891357},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5892999768257141},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.5169000029563904},{"id":"https://openalex.org/C49937458","wikidata":"https://www.wikidata.org/wiki/Q2599292","display_name":"Probabilistic logic","level":2,"score":0.5127999782562256},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4153999984264374},{"id":"https://openalex.org/C26760741","wikidata":"https://www.wikidata.org/wiki/Q160402","display_name":"Perception","level":2,"score":0.376800000667572},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.3504999876022339},{"id":"https://openalex.org/C95623464","wikidata":"https://www.wikidata.org/wiki/Q1096149","display_name":"Classifier (UML)","level":2,"score":0.35010001063346863},{"id":"https://openalex.org/C169900460","wikidata":"https://www.wikidata.org/wiki/Q2200417","display_name":"Cognition","level":2,"score":0.34369999170303345},{"id":"https://openalex.org/C178253425","wikidata":"https://www.wikidata.org/wiki/Q162668","display_name":"Visual perception","level":3,"score":0.3158999979496002},{"id":"https://openalex.org/C197115733","wikidata":"https://www.wikidata.org/wiki/Q1003136","display_name":"Forcing (mathematics)","level":2,"score":0.3142000138759613},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.30320000648498535},{"id":"https://openalex.org/C2777508537","wikidata":"https://www.wikidata.org/wiki/Q7936620","display_name":"Visual reasoning","level":2,"score":0.2955999970436096},{"id":"https://openalex.org/C89288958","wikidata":"https://www.wikidata.org/wiki/Q7301504","display_name":"Reasoning system","level":2,"score":0.27559998631477356},{"id":"https://openalex.org/C40506919","wikidata":"https://www.wikidata.org/wiki/Q7452469","display_name":"Sequence learning","level":2,"score":0.274399995803833},{"id":"https://openalex.org/C51167844","wikidata":"https://www.wikidata.org/wiki/Q4422623","display_name":"Latent variable","level":2,"score":0.27300000190734863},{"id":"https://openalex.org/C7149132","wikidata":"https://www.wikidata.org/wiki/Q1377840","display_name":"Forgetting","level":2,"score":0.27219998836517334},{"id":"https://openalex.org/C27753989","wikidata":"https://www.wikidata.org/wiki/Q284885","display_name":"Superposition principle","level":2,"score":0.2549000084400177},{"id":"https://openalex.org/C112933361","wikidata":"https://www.wikidata.org/wiki/Q2845258","display_name":"Probabilistic latent semantic analysis","level":2,"score":0.25440001487731934},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.25380000472068787},{"id":"https://openalex.org/C22019652","wikidata":"https://www.wikidata.org/wiki/Q331309","display_name":"Overfitting","level":3,"score":0.2535000145435333}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2601.06803","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2601.06803","pdf_url":"https://arxiv.org/pdf/2601.06803","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,"listed_in":[],"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-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2601.06803","doi":"https://doi.org/10.48550/arxiv.2601.06803","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2601.06803","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,"listed_in":[],"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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2601.06803","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2601.06803","pdf_url":"https://arxiv.org/pdf/2601.06803","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,"listed_in":[],"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-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7123568614.pdf","grobid_xml":"https://content.openalex.org/works/W7123568614.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7153011030","https://openalex.org/W7160319041","https://openalex.org/W7163007618","https://openalex.org/W7161353660","https://openalex.org/W7160979862","https://openalex.org/W7163159613","https://openalex.org/W7160867944","https://openalex.org/W7141460672","https://openalex.org/W7160207210","https://openalex.org/W7172379612"],"abstract_inverted_index":{"While":[0],"Chain-of-Thought":[1],"empowers":[2],"Large":[3],"Vision-Language":[4],"Models":[5],"with":[6,79,154],"multi-step":[7],"reasoning,":[8],"explicit":[9],"textual":[10],"rationales":[11],"suffer":[12],"from":[13],"an":[14],"information":[15],"bandwidth":[16],"bottleneck,":[17],"where":[18],"continuous":[19],"visual":[20,60],"details":[21],"are":[22],"discarded":[23],"during":[24],"discrete":[25],"tokenization.":[26],"Recent":[27],"latent":[28,77,137],"reasoning":[29,138],"methods":[30],"attempt":[31],"to":[32,40,45,97,108,168],"address":[33],"this":[34,50],"challenge,":[35],"but":[36],"often":[37],"fall":[38],"prey":[39],"premature":[41],"semantic":[42],"collapse":[43],"due":[44],"rigid":[46],"autoregressive":[47],"objectives.":[48],"In":[49],"paper,":[51],"we":[52],"propose":[53],"Laser,":[54],"a":[55,71,80,90,99],"novel":[56],"paradigm":[57],"that":[58,131],"reformulates":[59],"deduction":[61],"via":[62,115,122],"Dynamic":[63],"Windowed":[64],"Alignment":[65],"Learning":[66],"(DWAL).":[67],"Instead":[68],"of":[69,84,102],"forcing":[70],"point-wise":[72],"prediction,":[73],"Laser":[74,112,132],"aligns":[75],"the":[76,95,141],"state":[78],"dynamic":[81],"validity":[82],"window":[83],"future":[85],"semantics.":[86],"This":[87],"mechanism":[88],"enforces":[89],"\"Forest-before-Trees\"":[91],"cognitive":[92],"hierarchy,":[93],"enabling":[94],"model":[96],"maintain":[98],"probabilistic":[100],"superposition":[101],"global":[103],"features":[104],"before":[105],"narrowing":[106],"down":[107],"local":[109],"details.":[110],"Crucially,":[111],"maintains":[113],"interpretability":[114],"decodable":[116],"trajectories":[117],"while":[118,164],"stabilizing":[119],"unconstrained":[120],"learning":[121],"Self-Refined":[123],"Superposition.":[124],"Extensive":[125],"experiments":[126],"on":[127,147],"6":[128],"benchmarks":[129],"demonstrate":[130],"achieves":[133,151],"state-of-the-art":[134],"performance":[135],"among":[136],"methods,":[139],"surpassing":[140],"strong":[142],"baseline":[143],"Monet":[144],"by":[145,160],"5.03%":[146],"average.":[148],"Notably,":[149],"it":[150],"these":[152],"gains":[153],"extreme":[155],"efficiency,":[156],"reducing":[157],"inference":[158],"tokens":[159],"more":[161],"than":[162],"97%,":[163],"demonstrating":[165],"robust":[166],"generalization":[167],"out-of-distribution":[169],"domains.":[170]},"counts_by_year":[],"updated_date":"2026-10-31T12:09:11.096373","created_date":"2026-01-14T00:00:00"}