{"id":"https://openalex.org/W7160242965","doi":"https://doi.org/10.1109/wacv61042.2026.00502","title":"ST-Think: How Multimodal Large Language Models Reason About 4D Worlds from Ego-Centric Videos","display_name":"ST-Think: How Multimodal Large Language Models Reason About 4D Worlds from Ego-Centric Videos","publication_year":2026,"publication_date":"2026-03-06","ids":{"openalex":"https://openalex.org/W7160242965","doi":"https://doi.org/10.1109/wacv61042.2026.00502"},"language":null,"primary_location":{"id":"doi:10.1109/wacv61042.2026.00502","is_oa":false,"landing_page_url":"https://doi.org/10.1109/wacv61042.2026.00502","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5111341363","display_name":"Peiran Wu","orcid":null},"institutions":[{"id":"https://openalex.org/I36234482","display_name":"University of Bristol","ror":"https://ror.org/0524sp257","country_code":"GB","type":"education","lineage":["https://openalex.org/I36234482"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Peiran Wu","raw_affiliation_strings":["University of Bristol"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Bristol","institution_ids":["https://openalex.org/I36234482"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135328681","display_name":"Yunze Liu","orcid":null},"institutions":[{"id":"https://openalex.org/I4210161078","display_name":"P-D Refractories (Germany)","ror":"https://ror.org/0571c2151","country_code":"DE","type":"company","lineage":["https://openalex.org/I4210132412","https://openalex.org/I4210161078"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Yunze Liu","raw_affiliation_strings":["Memories.ai Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Memories.ai Research","institution_ids":["https://openalex.org/I4210161078"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135287845","display_name":"Miao Liu","orcid":"https://orcid.org/0000-0002-9033-9301"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Miao Liu","raw_affiliation_strings":["Meta"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100399628","display_name":"Junxiao Shen","orcid":"https://orcid.org/0000-0002-1552-4689"},"institutions":[{"id":"https://openalex.org/I36234482","display_name":"University of Bristol","ror":"https://ror.org/0524sp257","country_code":"GB","type":"education","lineage":["https://openalex.org/I36234482"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Junxiao Shen","raw_affiliation_strings":["University of Bristol"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Bristol","institution_ids":["https://openalex.org/I36234482"]}]}],"institutions":[],"countries_distinct_count":2,"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":0.68126084,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"5174","last_page":"5183"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.8220999836921692,"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":0.8220999836921692,"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/T10709","display_name":"Social Robot Interaction and HRI","score":0.012400000356137753,"subfield":{"id":"https://openalex.org/subfields/3207","display_name":"Social Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.011900000274181366,"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/natural-language","display_name":"Natural language","score":0.32330000400543213},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.31439998745918274},{"id":"https://openalex.org/keywords/field","display_name":"Field (mathematics)","score":0.2953999936580658},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.2847999930381775}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6342999935150146},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.42080000042915344},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.40049999952316284},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.3443000018596649},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.32330000400543213},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.31439998745918274},{"id":"https://openalex.org/C9652623","wikidata":"https://www.wikidata.org/wiki/Q190109","display_name":"Field (mathematics)","level":2,"score":0.2953999936580658},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.28769999742507935},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.2847999930381775},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.24330000579357147}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/wacv61042.2026.00502","is_oa":false,"landing_page_url":"https://doi.org/10.1109/wacv61042.2026.00502","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":23,"referenced_works":["https://openalex.org/W1985238052","https://openalex.org/W2765716052","https://openalex.org/W2788106261","https://openalex.org/W2964220823","https://openalex.org/W3098028994","https://openalex.org/W3205786327","https://openalex.org/W4313031313","https://openalex.org/W4390874416","https://openalex.org/W4402670894","https://openalex.org/W4402671548","https://openalex.org/W4402716288","https://openalex.org/W4403998229","https://openalex.org/W4404783328","https://openalex.org/W4413144646","https://openalex.org/W4413145998","https://openalex.org/W4413146301","https://openalex.org/W4413147595","https://openalex.org/W4414281281","https://openalex.org/W4415797648","https://openalex.org/W4415800665","https://openalex.org/W4416034567","https://openalex.org/W7133224126","https://openalex.org/W7151748244"],"related_works":[],"abstract_inverted_index":{"Humans":[0],"excel":[1],"at":[2],"spatial-temporal":[3,33,72,129],"reasoning,":[4],"effortlessly":[5],"interpreting":[6],"dynamic":[7],"visual":[8],"events":[9],"from":[10,35],"an":[11,36],"egocentric":[12,37],"viewpoint.":[13],"However,":[14],"whether":[15],"multimodal":[16,32],"large":[17],"language":[18],"models":[19],"(MLLMs)":[20],"can":[21],"similarly":[22],"understand":[23],"the":[24],"4D":[25],"world":[26],"remains":[27],"uncertain.":[28],"This":[29],"paper":[30],"explores":[31],"reasoning":[34,45,82,130],"perspective,":[38],"aiming":[39],"to":[40],"equip":[41],"MLLMs":[42],"with":[43,102,113],"human-like":[44],"capabilities.":[46],"To":[47],"support":[48],"this":[49],"objective,":[50],"we":[51,75],"introduce":[52],"Ego-ST":[53,117],"Bench,":[54],"a":[55,80],"novel":[56],"benchmark":[57],"containing":[58],"over":[59],"5,000":[60],"question-answer":[61],"pairs":[62],"across":[63],"four":[64],"categories,":[65],"systematically":[66],"evaluating":[67],"spatial,":[68],"temporal,":[69],"and":[70,119,124],"integrated":[71],"reasoning.":[73],"Additionally,":[74],"propose":[76],"ST-R1":[77,120],"training":[78],"paradigm,":[79],"video-based":[81,128],"model":[83],"that":[84],"incorporates":[85],"reverse":[86],"thinking":[87],"into":[88],"its":[89],"reinforcement":[90,108],"learning":[91],"process,":[92],"significantly":[93],"enhancing":[94],"performance.":[95],"We":[96],"combine":[97],"long-chain-of-thought":[98],"(long-CoT)":[99],"supervised":[100],"fine-tuning":[101],"Group":[103],"Relative":[104],"Policy":[105],"Optimization":[106],"(GRPO)":[107],"learning,":[109],"achieving":[110],"notable":[111],"improvements":[112],"limited":[114],"high-quality":[115],"data.":[116],"Bench":[118],"provide":[121],"valuable":[122],"insights":[123],"resources":[125],"for":[126],"advancing":[127],"research.":[131]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-05-06T00:00:00"}