{"id":"https://openalex.org/W7208715359","doi":"https://doi.org/10.48550/arxiv.2609.03892","title":"GraFT: A Training-Free Framework for Spatial Reasoning in Multimodal Large Language Models via 3D Scene Graphs","display_name":"GraFT: A Training-Free Framework for Spatial Reasoning in Multimodal Large Language Models via 3D Scene Graphs","publication_year":2026,"publication_date":"2026-09-03","ids":{"openalex":"https://openalex.org/W7208715359","doi":"https://doi.org/10.48550/arxiv.2609.03892"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2609.03892","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.03892","pdf_url":"https://arxiv.org/pdf/2609.03892","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","license_id":"https://openalex.org/licenses/cc-by","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/2609.03892","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5005001105","display_name":"\u675c\u4fca\u6e05","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I2250955327","display_name":"Huawei Technologies (China)","ror":"https://ror.org/00cmhce21","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250955327"]},{"id":"https://openalex.org/I4210146936","display_name":"Huawei Technologies (United States)","ror":"https://ror.org/03jyqk712","country_code":"US","type":"company","lineage":["https://openalex.org/I2250955327","https://openalex.org/I4210146936"]},{"id":"https://openalex.org/I4210160618","display_name":"Huawei Technologies (United Kingdom)","ror":"https://ror.org/056gzgs71","country_code":"GB","type":"company","lineage":["https://openalex.org/I2250955327","https://openalex.org/I4210160618"]}],"countries":["CN","GB","US"],"is_corresponding":false,"raw_author_name":"Du, Junqing","raw_affiliation_strings":["Riemann Lab , Huawei Technologies"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Riemann Lab , Huawei Technologies","institution_ids":["https://openalex.org/I2250955327","https://openalex.org/I4210146936","https://openalex.org/I4210160618"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129720319","display_name":"Fernando Ropero","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I2250955327","display_name":"Huawei Technologies (China)","ror":"https://ror.org/00cmhce21","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250955327"]},{"id":"https://openalex.org/I4210146936","display_name":"Huawei Technologies (United States)","ror":"https://ror.org/03jyqk712","country_code":"US","type":"company","lineage":["https://openalex.org/I2250955327","https://openalex.org/I4210146936"]},{"id":"https://openalex.org/I4210160618","display_name":"Huawei Technologies (United Kingdom)","ror":"https://ror.org/056gzgs71","country_code":"GB","type":"company","lineage":["https://openalex.org/I2250955327","https://openalex.org/I4210160618"]}],"countries":["CN","GB","US"],"is_corresponding":false,"raw_author_name":"Ropero, Fernando","raw_affiliation_strings":["Riemann Lab , Huawei Technologies"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Riemann Lab , Huawei Technologies","institution_ids":["https://openalex.org/I2250955327","https://openalex.org/I4210146936","https://openalex.org/I4210160618"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5018391591","display_name":"Erkin T\u00fcrk\u00f6z","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I2250955327","display_name":"Huawei Technologies (China)","ror":"https://ror.org/00cmhce21","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250955327"]},{"id":"https://openalex.org/I4210146936","display_name":"Huawei Technologies (United States)","ror":"https://ror.org/03jyqk712","country_code":"US","type":"company","lineage":["https://openalex.org/I2250955327","https://openalex.org/I4210146936"]},{"id":"https://openalex.org/I4210160618","display_name":"Huawei Technologies (United Kingdom)","ror":"https://ror.org/056gzgs71","country_code":"GB","type":"company","lineage":["https://openalex.org/I2250955327","https://openalex.org/I4210160618"]}],"countries":["CN","GB","US"],"is_corresponding":false,"raw_author_name":"Turkoz, Erkin","raw_affiliation_strings":["Riemann Lab , Huawei Technologies"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Riemann Lab , Huawei Technologies","institution_ids":["https://openalex.org/I2250955327","https://openalex.org/I4210146936","https://openalex.org/I4210160618"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5124787326","display_name":"Yanfeng Zhang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I2250955327","display_name":"Huawei Technologies (China)","ror":"https://ror.org/00cmhce21","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250955327"]},{"id":"https://openalex.org/I4210146936","display_name":"Huawei Technologies (United States)","ror":"https://ror.org/03jyqk712","country_code":"US","type":"company","lineage":["https://openalex.org/I2250955327","https://openalex.org/I4210146936"]},{"id":"https://openalex.org/I4210160618","display_name":"Huawei Technologies (United Kingdom)","ror":"https://ror.org/056gzgs71","country_code":"GB","type":"company","lineage":["https://openalex.org/I2250955327","https://openalex.org/I4210160618"]}],"countries":["CN","GB","US"],"is_corresponding":false,"raw_author_name":"Zhang, Yanfeng","raw_affiliation_strings":["Riemann Lab , Huawei Technologies"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Riemann Lab , Huawei Technologies","institution_ids":["https://openalex.org/I2250955327","https://openalex.org/I4210146936","https://openalex.org/I4210160618"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5150025646","display_name":"Lu Liu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I2250955327","display_name":"Huawei Technologies (China)","ror":"https://ror.org/00cmhce21","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250955327"]},{"id":"https://openalex.org/I4210146936","display_name":"Huawei Technologies (United States)","ror":"https://ror.org/03jyqk712","country_code":"US","type":"company","lineage":["https://openalex.org/I2250955327","https://openalex.org/I4210146936"]},{"id":"https://openalex.org/I4210160618","display_name":"Huawei Technologies (United Kingdom)","ror":"https://ror.org/056gzgs71","country_code":"GB","type":"company","lineage":["https://openalex.org/I2250955327","https://openalex.org/I4210160618"]}],"countries":["CN","GB","US"],"is_corresponding":false,"raw_author_name":"Liu, Lu","raw_affiliation_strings":["Riemann Lab , Huawei Technologies"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Riemann Lab , Huawei Technologies","institution_ids":["https://openalex.org/I2250955327","https://openalex.org/I4210146936","https://openalex.org/I4210160618"]}]}],"institutions":[],"countries_distinct_count":3,"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":0.9984999895095825,"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":0.9984999895095825,"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":0.00019999999494757503,"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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.00019999999494757503,"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/spatial-reasoning","display_name":"spatial reasoning","score":0.800000011920929},{"id":"https://openalex.org/keywords/multimodal-large-language-models","display_name":"multimodal large language models","score":0.9269999861717224},{"id":"https://openalex.org/keywords/3d-scene-graphs","display_name":"3D scene graphs","score":0.9869999885559082},{"id":"https://openalex.org/keywords/training-free-framework","display_name":"training-free framework","score":0.8109999895095825},{"id":"https://openalex.org/keywords/geometric-reasoning","display_name":"geometric reasoning","score":0.48100000619888306},{"id":"https://openalex.org/keywords/solid-geometry","display_name":"solid geometry","score":0.5889999866485596}],"concepts":[{"id":"https://openalex.org/C155911833","wikidata":"https://www.wikidata.org/wiki/Q3817354","display_name":"Spatial intelligence","level":2,"score":0.791100025177002},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.694100022315979},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5835999846458435},{"id":"https://openalex.org/C2777508537","wikidata":"https://www.wikidata.org/wiki/Q7936620","display_name":"Visual reasoning","level":2,"score":0.5238000154495239},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.5005000233650208},{"id":"https://openalex.org/C176217482","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Metric (unit)","level":2,"score":0.48739999532699585},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.4738999903202057},{"id":"https://openalex.org/C3019007443","wikidata":"https://www.wikidata.org/wiki/Q568742","display_name":"3d model","level":2,"score":0.420199990272522},{"id":"https://openalex.org/C108882727","wikidata":"https://www.wikidata.org/wiki/Q2991685","display_name":"Solid modeling","level":2,"score":0.4056999981403351},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.39980000257492065},{"id":"https://openalex.org/C179372163","wikidata":"https://www.wikidata.org/wiki/Q1406181","display_name":"Scene graph","level":3,"score":0.38339999318122864},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.3797000050544739},{"id":"https://openalex.org/C2778662690","wikidata":"https://www.wikidata.org/wiki/Q3125339","display_name":"Spatial ability","level":3,"score":0.36719998717308044},{"id":"https://openalex.org/C27511587","wikidata":"https://www.wikidata.org/wiki/Q2178623","display_name":"Spatial relation","level":2,"score":0.3391000032424927},{"id":"https://openalex.org/C146849305","wikidata":"https://www.wikidata.org/wiki/Q370766","display_name":"Ground truth","level":2,"score":0.3357999920845032},{"id":"https://openalex.org/C159620131","wikidata":"https://www.wikidata.org/wiki/Q1938983","display_name":"Spatial analysis","level":2,"score":0.3312000036239624},{"id":"https://openalex.org/C2777897806","wikidata":"https://www.wikidata.org/wiki/Q568742","display_name":"3D modeling","level":2,"score":0.32679998874664307},{"id":"https://openalex.org/C194969405","wikidata":"https://www.wikidata.org/wiki/Q170519","display_name":"Virtual reality","level":2,"score":0.3149000108242035},{"id":"https://openalex.org/C2777371692","wikidata":"https://www.wikidata.org/wiki/Q2178611","display_name":"Spatial cognition","level":3,"score":0.296999990940094},{"id":"https://openalex.org/C89288958","wikidata":"https://www.wikidata.org/wiki/Q7301504","display_name":"Reasoning system","level":2,"score":0.29429998993873596},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.2912999987602234},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.2793000042438507},{"id":"https://openalex.org/C83725634","wikidata":"https://www.wikidata.org/wiki/Q7268699","display_name":"Qualitative reasoning","level":2,"score":0.2786000072956085},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.2750000059604645},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.27300000190734863},{"id":"https://openalex.org/C162319229","wikidata":"https://www.wikidata.org/wiki/Q175263","display_name":"Data structure","level":2,"score":0.2574000060558319},{"id":"https://openalex.org/C88230418","wikidata":"https://www.wikidata.org/wiki/Q131476","display_name":"Graph theory","level":2,"score":0.25360000133514404}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2609.03892","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.03892","pdf_url":"https://arxiv.org/pdf/2609.03892","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","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2609.03892","doi":"https://doi.org/10.48550/arxiv.2609.03892","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2609.03892","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2609.03892","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.03892","pdf_url":"https://arxiv.org/pdf/2609.03892","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","license_id":"https://openalex.org/licenses/cc-by","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/W7208715359.pdf","grobid_xml":"https://content.openalex.org/works/W7208715359.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7134808437","https://openalex.org/W4416159329","https://openalex.org/W7152720431","https://openalex.org/W7160905493","https://openalex.org/W7133351226","https://openalex.org/W7167786781","https://openalex.org/W7140283259","https://openalex.org/W7133305662","https://openalex.org/W4416998161","https://openalex.org/W7168244276"],"abstract_inverted_index":{"3D":[0,58,83,90,98],"spatial":[1,165],"reasoning":[2],"underpins":[3],"understanding":[4],"and":[5,33,36,68,123,157,161],"acting":[6],"in":[7,15,104],"the":[8,46,63,81,96,101,117,127,137],"physical":[9],"world,":[10],"yet":[11],"it":[12],"remains":[13],"unreliable":[14],"current":[16],"multimodal":[17],"large":[18],"language":[19],"models":[20,23],"(MLLMs).":[21],"These":[22],"falter":[24],"at":[25,29,37],"precise":[26],"geometric":[27],"measurement,":[28],"transforming":[30],"between":[31],"egocentric":[32,125],"allocentric":[34],"viewpoints,":[35],"grounding":[38],"fine-grained":[39],"appearance.":[40],"The":[41],"most":[42],"common":[43],"remedies":[44],"fine-tune":[45],"model":[47,103,118],"on":[48],"large-scale":[49],"curated":[50],"spatial-reasoning":[51],"datasets":[52],"or":[53],"attach":[54],"dedicated":[55],"encoders":[56],"for":[57],"geometry,":[59],"which":[60],"typically":[61],"couples":[62],"solution":[64],"to":[65,100,152],"costly":[66],"supervision":[67],"a":[69,76,86,114],"specific":[70],"backbone.":[71],"We":[72],"instead":[73],"introduce":[74],"GraFT,":[75],"training-free":[77],"framework":[78],"that":[79],"supplies":[80,95],"missing":[82],"structure":[84,99],"through":[85],"compact,":[87],"easily":[88],"maintained":[89],"scene":[91],"graph":[92],"(3DSG).":[93],"GraFT":[94,132,146],"graph's":[97],"frozen":[102,148],"three":[105],"forms:":[106],"as":[107,113,120,124],"values":[108],"computed":[109],"by":[110,142,150],"symbolic":[111],"tools,":[112],"bird's-eye-view":[115],"(BEV)":[116],"reads":[119],"an":[121],"image,":[122],"frames":[126],"geometry":[128],"selects.":[129],"On":[130,144],"ScanQA,":[131],"improves":[133,147],"every":[134,155],"metric":[135],"over":[136],"same-backbone":[138],"baseline,":[139,160],"raising":[140],"CIDEr":[141],"27%.":[143],"VSI-Bench,":[145],"MLLMs":[149],"up":[151],"65%,":[153],"surpassing":[154],"proprietary":[156],"general-purpose":[158],"open-source":[159],"several":[162],"prominent":[163],"fine-tuned":[164],"models.":[166]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-09-05T00:00:00"}
