{"id":"https://openalex.org/W4415320074","doi":"https://doi.org/10.48550/arxiv.2504.01805","title":"SpaceR: Reinforcing MLLMs in Video Spatial Reasoning","display_name":"SpaceR: Reinforcing MLLMs in Video Spatial Reasoning","publication_year":2025,"publication_date":"2025-04-02","ids":{"openalex":"https://openalex.org/W4415320074","doi":"https://doi.org/10.48550/arxiv.2504.01805"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2504.01805","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2504.01805","pdf_url":"https://arxiv.org/pdf/2504.01805","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","license_id":"https://openalex.org/licenses/cc-by-nc","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2504.01805","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5120051052","display_name":"Kun Ouyang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ouyang, Kun","raw_affiliation_strings":["National Key Laboratory for Multimedia Information Processing , School of Computer Science , Peking University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"National Key Laboratory for Multimedia Information Processing , School of Computer Science , Peking University","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5044541656","display_name":"Yuanxin Liu","orcid":"https://orcid.org/0000-0002-0673-4887","observed_orcids":["https://orcid.org/0000-0002-0673-4887"]},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Liu, Yuanxin","raw_affiliation_strings":["National Key Laboratory for Multimedia Information Processing , School of Computer Science , Peking University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"National Key Laboratory for Multimedia Information Processing , School of Computer Science , Peking University","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5039135087","display_name":"Haoning Wu","orcid":"https://orcid.org/0009-0001-8717-338X","observed_orcids":["https://orcid.org/0009-0001-8717-338X"]},"institutions":[{"id":"https://openalex.org/I172675005","display_name":"Nanyang Technological University","ror":"https://ror.org/02e7b5302","country_code":"SG","type":"education","lineage":["https://openalex.org/I172675005"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Wu, Haoning","raw_affiliation_strings":["Nanyang Technological University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Nanyang Technological University","institution_ids":["https://openalex.org/I172675005"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5115883506","display_name":"Yi Liu","orcid":"https://orcid.org/0000-0002-4978-127X","observed_orcids":["https://orcid.org/0000-0002-4978-127X","https://orcid.org/0009-0007-8292-3296"]},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Liu, Yi","raw_affiliation_strings":["National Key Laboratory for Multimedia Information Processing , School of Computer Science , Peking University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"National Key Laboratory for Multimedia Information Processing , School of Computer Science , Peking University","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Zhou, Hao","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhou, Hao","raw_affiliation_strings":["WeChat AI , Tencent Inc. , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"WeChat AI , Tencent Inc. , China","institution_ids":["https://openalex.org/I2250653659"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5109296925","display_name":"Jie Zhou","orcid":"https://orcid.org/0009-0006-2582-0867","observed_orcids":["https://orcid.org/0009-0006-2582-0867"]},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhou, Jie","raw_affiliation_strings":["WeChat AI , Tencent Inc. , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"WeChat AI , Tencent Inc. , China","institution_ids":["https://openalex.org/I2250653659"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5024849044","display_name":"Fandong Meng","orcid":"https://orcid.org/0000-0002-8158-2377","observed_orcids":["https://orcid.org/0000-0002-8158-2377"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Meng, Fandong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5112082146","display_name":"Xu Guang Sun","orcid":"https://orcid.org/0009-0000-5404-9246","observed_orcids":["https://orcid.org/0009-0000-5404-9246"]},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Sun, Xu","raw_affiliation_strings":["National Key Laboratory for Multimedia Information Processing , School of Computer Science , Peking University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"National Key Laboratory for Multimedia Information Processing , School of Computer Science , Peking University","institution_ids":["https://openalex.org/I20231570"]}]}],"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":162,"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/T10653","display_name":"Robot Manipulation and Learning","score":0.0,"subfield":{"id":"https://openalex.org/subfields/2207","display_name":"Control and Systems Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"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"}}],"keywords":[{"id":"https://openalex.org/keywords/multimodal-large-language-models","display_name":"multimodal large language models","score":0.9929999709129333},{"id":"https://openalex.org/keywords/reinforcement-learning-with-verifiable-rewards","display_name":"reinforcement learning with verifiable rewards","score":0.9449999928474426},{"id":"https://openalex.org/keywords/group-relative-policy-optimization","display_name":"Group Relative Policy Optimization","score":0.8130000233650208},{"id":"https://openalex.org/keywords/video-understanding","display_name":"video understanding","score":0.7009999752044678}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7468000054359436},{"id":"https://openalex.org/C155911833","wikidata":"https://www.wikidata.org/wiki/Q3817354","display_name":"Spatial intelligence","level":2,"score":0.6987000107765198},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6297000050544739},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.5647000074386597},{"id":"https://openalex.org/C27511587","wikidata":"https://www.wikidata.org/wiki/Q2178623","display_name":"Spatial relation","level":2,"score":0.48330000042915344},{"id":"https://openalex.org/C85847156","wikidata":"https://www.wikidata.org/wiki/Q59015987","display_name":"Verifiable secret sharing","level":3,"score":0.4819999933242798},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4700999855995178},{"id":"https://openalex.org/C64754055","wikidata":"https://www.wikidata.org/wiki/Q7574053","display_name":"Spatial contextual awareness","level":2,"score":0.3564000129699707},{"id":"https://openalex.org/C159620131","wikidata":"https://www.wikidata.org/wiki/Q1938983","display_name":"Spatial analysis","level":2,"score":0.35569998621940613},{"id":"https://openalex.org/C22367795","wikidata":"https://www.wikidata.org/wiki/Q7625208","display_name":"Structured prediction","level":2,"score":0.3010999858379364},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.2944999933242798},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.2921000123023987},{"id":"https://openalex.org/C37335422","wikidata":"https://www.wikidata.org/wiki/Q6888134","display_name":"Model-based reasoning","level":3,"score":0.2614000141620636},{"id":"https://openalex.org/C2778662690","wikidata":"https://www.wikidata.org/wiki/Q3125339","display_name":"Spatial ability","level":3,"score":0.25220000743865967}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2504.01805","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2504.01805","pdf_url":"https://arxiv.org/pdf/2504.01805","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","license_id":"https://openalex.org/licenses/cc-by-nc","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2504.01805","doi":"https://doi.org/10.48550/arxiv.2504.01805","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2504.01805","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:2504.01805","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2504.01805","pdf_url":"https://arxiv.org/pdf/2504.01805","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","license_id":"https://openalex.org/licenses/cc-by-nc","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4415320074.pdf","grobid_xml":"https://content.openalex.org/works/W4415320074.grobid-xml"},"referenced_works_count":23,"referenced_works":["https://openalex.org/W4391631327","https://openalex.org/W4391833450","https://openalex.org/W4402727142","https://openalex.org/W4403370767","https://openalex.org/W4404330874","https://openalex.org/W4405172832","https://openalex.org/W4405655184","https://openalex.org/W4405766390","https://openalex.org/W4406093553","https://openalex.org/W4406489046","https://openalex.org/W4406774573","https://openalex.org/W4406780478","https://openalex.org/W4407794264","https://openalex.org/W4409316930","https://openalex.org/W4413144646","https://openalex.org/W4413146669","https://openalex.org/W4414281281","https://openalex.org/W4414565117","https://openalex.org/W4414846785","https://openalex.org/W4415063999","https://openalex.org/W4415085924","https://openalex.org/W4415308770","https://openalex.org/W4417250113"],"related_works":["https://openalex.org/W7124297557","https://openalex.org/W7090931384","https://openalex.org/W7160905493","https://openalex.org/W7169871736","https://openalex.org/W4415063999","https://openalex.org/W4413144646","https://openalex.org/W4417228662","https://openalex.org/W7138367878","https://openalex.org/W6891934496","https://openalex.org/W7140317515"],"abstract_inverted_index":{"Video":[0],"spatial":[1,8,49,76,101,145,155,166,219],"reasoning,":[2],"which":[3,139],"involves":[4],"inferring":[5],"the":[6,31,41,54,79,87,141,148,190,203,209],"underlying":[7],"structure":[9],"from":[10,29],"observed":[11],"video":[12,75,179],"frames,":[13],"poses":[14],"a":[15,94,121,134],"significant":[16],"challenge":[17],"for":[18,36,110],"existing":[19],"Multimodal":[20],"Large":[21],"Language":[22],"Models":[23],"(MLLMs).":[24],"This":[25],"limitation":[26],"stems":[27],"primarily":[28],"1)":[30],"absence":[32],"of":[33,43,56,211,222],"high-quality":[34],"datasets":[35],"this":[37,68,83],"task,":[38],"and":[39,107,172,185,198,215,226],"2)":[40],"lack":[42],"effective":[44,154],"training":[45],"strategies":[46],"to":[47,71,143],"develop":[48],"reasoning":[50,66,77,102,167,220],"capabilities.":[51],"Motivated":[52],"by":[53,193],"success":[55],"Reinforcement":[57],"Learning":[58],"with":[59,96,104,133,202],"Verifiable":[60],"Reward":[61],"(RLVR)":[62],"in":[63,74,147,217],"unlocking":[64],"LLM":[65],"abilities,":[67],"work":[69],"aims":[70],"improve":[72],"MLLMs":[73],"through":[78],"RLVR":[80,119],"paradigm.":[81],"To":[82],"end,":[84],"we":[85,91,116],"introduce":[86],"$\\textbf{SpaceR}$":[88],"framework.":[89],"First,":[90],"present":[92],"$\\textbf{SpaceR-151k}$,":[93],"dataset":[95,214,227],"91k":[97],"questions":[98],"spanning":[99],"diverse":[100],"scenarios":[103],"verifiable":[105],"answers,":[106],"60k":[108],"samples":[109],"maintaining":[111,175],"general":[112],"multimodal":[113],"understanding.":[114],"Second,":[115],"propose":[117],"$\\textbf{Spatially-Guided":[118],"(SG-RLVR)}$,":[120],"novel":[122,135],"reinforcement":[123],"learning":[124],"approach":[125],"that":[126,160],"extends":[127],"Group":[128],"Relative":[129],"Policy":[130],"Optimization":[131],"(GRPO)":[132],"map":[136],"imagination":[137],"mechanism,":[138],"encourages":[140],"model":[142,206],"infer":[144],"layouts":[146],"thinking":[149],"process,":[150],"thereby":[151],"facilitating":[152],"more":[153],"reasoning.":[156],"Extensive":[157],"experiments":[158],"demonstrate":[159],"SpaceR":[161,188],"achieves":[162],"state-of-the-art":[163],"performance":[164],"on":[165,178,196,200],"benchmarks":[168,181],"(e.g.,":[169,182],"VSI-Bench,":[170],"STI-Bench,":[171],"SPAR-Bench),":[173],"while":[174],"competitive":[176],"results":[177],"understanding":[180],"Video-MME,":[183],"TempCompass,":[184],"LongVideoBench).":[186],"Remarkably,":[187],"surpasses":[189],"advanced":[191],"GPT-4o":[192],"11.6\\%":[194],"accuracy":[195],"VSI-Bench":[197],"is":[199],"par":[201],"leading":[204],"proprietary":[205],"Gemini-2.0-Flash,":[207],"highlighting":[208],"effectiveness":[210],"our":[212],"SpaceR-151k":[213],"SG-RLVR":[216],"reinforcing":[218],"ability":[221],"MLLMs.":[223],"Code,":[224],"model,":[225],"are":[228],"available":[229],"at":[230],"https://github.com/OuyangKun10/SpaceR.":[231]},"counts_by_year":[{"year":2026,"cited_by_count":123},{"year":2025,"cited_by_count":39}],"updated_date":"2026-11-09T13:22:03.289127","created_date":"2025-10-18T00:00:00"}