{"id":"https://openalex.org/W7220927465","doi":"https://doi.org/10.1109/cvprf71909.2026.00908","title":"Mull-Tokens: Modality-Agnostic Latent Thinking","display_name":"Mull-Tokens: Modality-Agnostic Latent Thinking","publication_year":2026,"publication_date":"2026-06-01","ids":{"openalex":"https://openalex.org/W7220927465","doi":"https://doi.org/10.1109/cvprf71909.2026.00908"},"language":"en","primary_location":{"id":"doi:10.1109/cvprf71909.2026.00908","doi":"https://doi.org/10.1109/cvprf71909.2026.00908","is_oa":true,"landing_page_url":"https://doi.org/10.1109/cvprf71909.2026.00908","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE/CVF Conference on Computer Vision and Pattern Recognition- FINDINGS Track (CVPRF)","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","is_traditional":true,"indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1109/cvprf71909.2026.00908","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5065379518","display_name":"Arijit Ray","orcid":"https://orcid.org/0000-0002-4175-0655","observed_orcids":["https://orcid.org/0000-0002-4175-0655"]},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Arijit Ray","raw_affiliation_strings":["Google"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100712178","display_name":"Ahmed Abdelkader","orcid":"https://orcid.org/0000-0002-6749-1807","observed_orcids":["https://orcid.org/0000-0002-6749-1807","https://orcid.org/0009-0004-2797-702X"]},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ahmed Abdelkader","raw_affiliation_strings":["Google"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5159279310","display_name":"Chengzhi Mao","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Chengzhi Mao","raw_affiliation_strings":["Google"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5061227594","display_name":"Bryan A. Plummer","orcid":"https://orcid.org/0000-0002-7074-3219","observed_orcids":["https://orcid.org/0000-0002-7074-3219"]},"institutions":[{"id":"https://openalex.org/I111088046","display_name":"Boston University","ror":"https://ror.org/05qwgg493","country_code":"US","type":"education","lineage":["https://openalex.org/I111088046"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Bryan A. Plummer","raw_affiliation_strings":["Boston University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Boston University","institution_ids":["https://openalex.org/I111088046"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5075906727","display_name":"Kate Saenko","orcid":"https://orcid.org/0000-0002-7564-7218","observed_orcids":["https://orcid.org/0000-0002-7564-7218","https://orcid.org/0000-0002-5704-7614"]},"institutions":[{"id":"https://openalex.org/I111088046","display_name":"Boston University","ror":"https://ror.org/05qwgg493","country_code":"US","type":"education","lineage":["https://openalex.org/I111088046"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kate Saenko","raw_affiliation_strings":["Boston University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Boston University","institution_ids":["https://openalex.org/I111088046"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5032451496","display_name":"Ranjay Krishna","orcid":"https://orcid.org/0000-0001-8784-2531","observed_orcids":["https://orcid.org/0000-0001-8784-2531"]},"institutions":[{"id":"https://openalex.org/I201448701","display_name":"University of Washington","ror":"https://ror.org/00cvxb145","country_code":"US","type":"education","lineage":["https://openalex.org/I201448701"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ranjay Krishna","raw_affiliation_strings":["University of Washington"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Washington","institution_ids":["https://openalex.org/I201448701"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5065368881","display_name":"Leonidas Guibas","orcid":"https://orcid.org/0000-0002-8315-4886","observed_orcids":["https://orcid.org/0000-0002-8315-4886"]},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Leonidas Guibas","raw_affiliation_strings":["Google"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5077732223","display_name":"Wen-Sheng Chu","orcid":"https://orcid.org/0000-0001-8592-6088","observed_orcids":["https://orcid.org/0000-0001-8592-6088"]},"institutions":[{"id":"https://openalex.org/I97018004","display_name":"Stanford University","ror":"https://ror.org/00f54p054","country_code":"US","type":"education","lineage":["https://openalex.org/I97018004"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Wen-Sheng Chu","raw_affiliation_strings":["Stanford University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Stanford University","institution_ids":["https://openalex.org/I97018004"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.7289816,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"9478","last_page":"9489"},"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/T10028","display_name":"Topic Modeling","score":0.0,"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/T11010","display_name":"Logic, Reasoning, and Knowledge","score":0.0,"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/multimodal-reasoning","display_name":"multimodal reasoning","score":0.9110000133514404},{"id":"https://openalex.org/keywords/spatial-reasoning","display_name":"spatial reasoning","score":0.6620000004768372},{"id":"https://openalex.org/keywords/pretraining","display_name":"pre-training","score":0.36500000953674316},{"id":"https://openalex.org/keywords/puzzle-solving","display_name":"puzzle solving","score":0.8379999995231628},{"id":"https://openalex.org/keywords/perspective-taking","display_name":"perspective taking","score":0.45500001311302185},{"id":"https://openalex.org/keywords/visual-grounding","display_name":"visual grounding","score":0.5450000166893005}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6685000061988831},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5939000248908997},{"id":"https://openalex.org/C103057564","wikidata":"https://www.wikidata.org/wiki/Q4751139","display_name":"Analytic reasoning","level":3,"score":0.5576000213623047},{"id":"https://openalex.org/C2777508537","wikidata":"https://www.wikidata.org/wiki/Q7936620","display_name":"Visual reasoning","level":2,"score":0.555899977684021},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.49889999628067017},{"id":"https://openalex.org/C183521366","wikidata":"https://www.wikidata.org/wiki/Q7256422","display_name":"Psychology of reasoning","level":4,"score":0.45170000195503235},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4004000127315521},{"id":"https://openalex.org/C89288958","wikidata":"https://www.wikidata.org/wiki/Q7301504","display_name":"Reasoning system","level":2,"score":0.400299996137619},{"id":"https://openalex.org/C2779903281","wikidata":"https://www.wikidata.org/wiki/Q6888026","display_name":"Modalities","level":2,"score":0.39899998903274536},{"id":"https://openalex.org/C115086926","wikidata":"https://www.wikidata.org/wiki/Q17004651","display_name":"Causal reasoning","level":3,"score":0.388700008392334},{"id":"https://openalex.org/C180747234","wikidata":"https://www.wikidata.org/wiki/Q23373","display_name":"Cognitive psychology","level":1,"score":0.3666999936103821},{"id":"https://openalex.org/C2776502983","wikidata":"https://www.wikidata.org/wiki/Q690182","display_name":"Contrast (vision)","level":2,"score":0.3659000098705292},{"id":"https://openalex.org/C155911833","wikidata":"https://www.wikidata.org/wiki/Q3817354","display_name":"Spatial intelligence","level":2,"score":0.35429999232292175},{"id":"https://openalex.org/C83725634","wikidata":"https://www.wikidata.org/wiki/Q7268699","display_name":"Qualitative reasoning","level":2,"score":0.3319999873638153},{"id":"https://openalex.org/C86827895","wikidata":"https://www.wikidata.org/wiki/Q7098582","display_name":"Opportunistic reasoning","level":4,"score":0.3296999931335449},{"id":"https://openalex.org/C36964233","wikidata":"https://www.wikidata.org/wiki/Q7920942","display_name":"Verbal reasoning","level":3,"score":0.32249999046325684},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.313400000333786},{"id":"https://openalex.org/C20162079","wikidata":"https://www.wikidata.org/wiki/Q1151406","display_name":"Case-based reasoning","level":2,"score":0.3125999867916107},{"id":"https://openalex.org/C188147891","wikidata":"https://www.wikidata.org/wiki/Q147638","display_name":"Cognitive science","level":1,"score":0.2906999886035919},{"id":"https://openalex.org/C37335422","wikidata":"https://www.wikidata.org/wiki/Q6888134","display_name":"Model-based reasoning","level":3,"score":0.2888999879360199},{"id":"https://openalex.org/C59656382","wikidata":"https://www.wikidata.org/wiki/Q191536","display_name":"Conjunction (astronomy)","level":2,"score":0.28200000524520874},{"id":"https://openalex.org/C51167844","wikidata":"https://www.wikidata.org/wiki/Q4422623","display_name":"Latent variable","level":2,"score":0.27709999680519104},{"id":"https://openalex.org/C159032336","wikidata":"https://www.wikidata.org/wiki/Q2488768","display_name":"Non-monotonic logic","level":2,"score":0.2736000120639801},{"id":"https://openalex.org/C107848011","wikidata":"https://www.wikidata.org/wiki/Q4680756","display_name":"Adaptive reasoning","level":4,"score":0.27079999446868896},{"id":"https://openalex.org/C97364631","wikidata":"https://www.wikidata.org/wiki/Q484284","display_name":"Deductive reasoning","level":2,"score":0.2619999945163727},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2533999979496002}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/cvprf71909.2026.00908","doi":"https://doi.org/10.1109/cvprf71909.2026.00908","is_oa":true,"landing_page_url":"https://doi.org/10.1109/cvprf71909.2026.00908","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE/CVF Conference on Computer Vision and Pattern Recognition- FINDINGS Track (CVPRF)","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1109/cvprf71909.2026.00908","doi":"https://doi.org/10.1109/cvprf71909.2026.00908","is_oa":true,"landing_page_url":"https://doi.org/10.1109/cvprf71909.2026.00908","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE/CVF Conference on Computer Vision and Pattern Recognition- FINDINGS Track (CVPRF)","raw_type":"proceedings-article"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":26,"referenced_works":["https://openalex.org/W2103679996","https://openalex.org/W2406512702","https://openalex.org/W3081168214","https://openalex.org/W4239441472","https://openalex.org/W4403923795","https://openalex.org/W4407310640","https://openalex.org/W4413144646","https://openalex.org/W4413145497","https://openalex.org/W4413146707","https://openalex.org/W4413156213","https://openalex.org/W4414281281","https://openalex.org/W4414581417","https://openalex.org/W4415063999","https://openalex.org/W4415796658","https://openalex.org/W4415800507","https://openalex.org/W4415800665","https://openalex.org/W4416031380","https://openalex.org/W4416037467","https://openalex.org/W4416082444","https://openalex.org/W4416284578","https://openalex.org/W7133193597","https://openalex.org/W7133198832","https://openalex.org/W7133208214","https://openalex.org/W7133224126","https://openalex.org/W7160094696","https://openalex.org/W7196993360"],"related_works":["https://openalex.org/W2499373164","https://openalex.org/W7221032939","https://openalex.org/W4402684048","https://openalex.org/W6935691267","https://openalex.org/W6903830402","https://openalex.org/W7210263777","https://openalex.org/W6922619513","https://openalex.org/W7221029504","https://openalex.org/W7139934461","https://openalex.org/W6922430350"],"abstract_inverted_index":{"Reasoning":[0],"goes":[1],"beyond":[2],"language;":[3],"the":[4,25,79,113],"real":[5],"world":[6],"requires":[7],"reasoning":[8,28,92,120,142],"about":[9],"space,":[10],"time,":[11],"affordances,":[12],"and":[13,33,104,128,152,175],"much":[14],"more":[15],"that":[16,60,134],"words":[17],"alone":[18],"cannot":[19],"convey.":[20],"Existing":[21],"multimodal":[22],"models":[23],"exploring":[24],"potential":[26],"of":[27,45],"with":[29],"images":[30],"are":[31],"brittle":[32],"do":[34],"not":[35],"scale.":[36],"They":[37],"rely":[38],"on":[39,156],"calling":[40],"specialist":[41],"tools,":[42],"costly":[43],"generation":[44],"images,":[46],"or":[47,71,143],"handcrafted":[48],"datasets.":[49],"We":[50,82,94],"offer":[51],"a":[52,148,157,180],"simpler":[53],"alternative":[54],"-":[55,57],"Mull-Tokens":[56,88,97,135,178],"latent":[58,91],"tokens":[59],"can":[61],"be":[62],"pre-trained":[63],"to":[64,76,86,154,163,168,183],"hold":[65],"intermediate":[66],"information":[67],"in":[68,172,186],"either":[69],"image":[70],"text":[72],"modalities":[73],"so":[74],"as":[75,125],"think":[77,185],"towards":[78],"correct":[80],"answer.":[81],"investigate":[83],"best":[84],"practices":[85],"train":[87,96],"inspired":[89],"by":[90,110],"frameworks.":[93],"first":[95],"using":[98,112],"supervision":[99,109],"from":[100],"interleaved":[101,144],"text-image":[102],"traces,":[103],"then":[105],"finetune":[106],"without":[107],"any":[108],"only":[111],"final":[114],"answers.":[115],"Across":[116],"four":[117],"challenging":[118],"spatial":[119],"benchmarks":[121],"involving":[122],"tasks":[123],"such":[124],"solving":[126,159],"puzzles":[127],"taking":[129],"different":[130],"perspectives,":[131],"we":[132],"demonstrate":[133],"improve":[136],"upon":[137],"several":[138],"baselines":[139],"utilizing":[140],"text-only":[141],"image-text":[145],"reasoning,":[146,177],"achieving":[147],"+3%":[149],"average":[150],"improvement":[151],"up":[153],"+16%":[155],"puzzle":[158],"reasoning-heavy":[160],"split":[161],"compared":[162],"our":[164],"strongest":[165],"baseline.":[166],"Adding":[167],"conversations":[169],"around":[170],"challenges":[171],"grounding":[173],"textual":[174],"visual":[176],"offers":[179],"simple":[181],"solution":[182],"implicitly":[184],"multiple":[187],"modalities.":[188]},"counts_by_year":[],"updated_date":"2026-11-08T12:12:28.789561","created_date":"2026-10-09T00:00:00"}