{"id":"https://openalex.org/W7204267973","doi":"https://doi.org/10.48550/arxiv.2608.24574","title":"PhysMLLMs: Spatial Priors for Unified Referring Segmentation and Grounded Reasoning of Images and Videos","display_name":"PhysMLLMs: Spatial Priors for Unified Referring Segmentation and Grounded Reasoning of Images and Videos","publication_year":2026,"publication_date":"2026-08-25","ids":{"openalex":"https://openalex.org/W7204267973","doi":"https://doi.org/10.48550/arxiv.2608.24574"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2608.24574","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.24574","pdf_url":"https://arxiv.org/pdf/2608.24574","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":"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/2608.24574","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5148218656","display_name":"Siyao Yan","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I76214153","display_name":"Lanzhou University","ror":"https://ror.org/01mkqqe32","country_code":"CN","type":"education","lineage":["https://openalex.org/I76214153"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yan, Siyao","raw_affiliation_strings":["School of Information Science and Engineering , Lanzhou University , Lanzhou , 730000 , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Information Science and Engineering , Lanzhou University , Lanzhou , 730000 , China","institution_ids":["https://openalex.org/I76214153"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5126779638","display_name":"Bo Han","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I76214153","display_name":"Lanzhou University","ror":"https://ror.org/01mkqqe32","country_code":"CN","type":"education","lineage":["https://openalex.org/I76214153"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Han, Bo","raw_affiliation_strings":["School of Information Science and Engineering , Lanzhou University , Lanzhou , 730000 , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Information Science and Engineering , Lanzhou University , Lanzhou , 730000 , China","institution_ids":["https://openalex.org/I76214153"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5058458910","display_name":"Jisheng Dang","orcid":"https://orcid.org/0000-0002-5378-6225","observed_orcids":["https://orcid.org/0000-0002-5378-6225","https://orcid.org/0000-0002-5513-911X","https://orcid.org/0009-0006-0082-1935"]},"institutions":[{"id":"https://openalex.org/I76214153","display_name":"Lanzhou University","ror":"https://ror.org/01mkqqe32","country_code":"CN","type":"education","lineage":["https://openalex.org/I76214153"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Dang, Jisheng","raw_affiliation_strings":["SCIENCE CHINA Information Sciences ","School of Information Science and Engineering , Lanzhou University , Lanzhou , 730000 , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"SCIENCE CHINA Information Sciences ","institution_ids":[]},{"raw_affiliation_string":"School of Information Science and Engineering , Lanzhou University , Lanzhou , 730000 , China","institution_ids":["https://openalex.org/I76214153"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138426330","display_name":"Bimei Wang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I76214153","display_name":"Lanzhou University","ror":"https://ror.org/01mkqqe32","country_code":"CN","type":"education","lineage":["https://openalex.org/I76214153"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wang, Bimei","raw_affiliation_strings":["School of Information Science and Engineering , Lanzhou University , Lanzhou , 730000 , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Information Science and Engineering , Lanzhou University , Lanzhou , 730000 , China","institution_ids":["https://openalex.org/I76214153"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5148248076","display_name":"Shude Wang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I76214153","display_name":"Lanzhou University","ror":"https://ror.org/01mkqqe32","country_code":"CN","type":"education","lineage":["https://openalex.org/I76214153"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wang, Shude","raw_affiliation_strings":["School of Information Science and Engineering , Lanzhou University , Lanzhou , 730000 , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Information Science and Engineering , Lanzhou University , Lanzhou , 730000 , China","institution_ids":["https://openalex.org/I76214153"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5122980723","display_name":"Hong Peng","orcid":"https://orcid.org/0000-0003-1558-1269","observed_orcids":["https://orcid.org/0000-0003-1558-1269"]},"institutions":[{"id":"https://openalex.org/I76214153","display_name":"Lanzhou University","ror":"https://ror.org/01mkqqe32","country_code":"CN","type":"education","lineage":["https://openalex.org/I76214153"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Peng, Hong","raw_affiliation_strings":["School of Information Science and Engineering , Lanzhou University , Lanzhou , 730000 , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Information Science and Engineering , Lanzhou University , Lanzhou , 730000 , China","institution_ids":["https://openalex.org/I76214153"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133182800","display_name":"Yulan Guo","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I157773358","display_name":"Sun Yat-sen University","ror":"https://ror.org/0064kty71","country_code":"CN","type":"education","lineage":["https://openalex.org/I157773358"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Guo, Yulan","raw_affiliation_strings":["School of Electronics and Communication Engineering , Sun Yat-sen University , Shenzhen 510006 , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Electronics and Communication Engineering , Sun Yat-sen University , Shenzhen 510006 , China","institution_ids":["https://openalex.org/I157773358"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5148403087","display_name":"Jianhuang Lai","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I157773358","display_name":"Sun Yat-sen University","ror":"https://ror.org/0064kty71","country_code":"CN","type":"education","lineage":["https://openalex.org/I157773358"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Lai, Jianhuang","raw_affiliation_strings":["School of Electronics and Communication Engineering , Sun Yat-sen University , Shenzhen 510006 , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Electronics and Communication Engineering , Sun Yat-sen University , Shenzhen 510006 , China","institution_ids":["https://openalex.org/I157773358"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138755689","display_name":"Bin Hu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I76214153","display_name":"Lanzhou University","ror":"https://ror.org/01mkqqe32","country_code":"CN","type":"education","lineage":["https://openalex.org/I76214153"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hu, Bin","raw_affiliation_strings":["School of Information Science and Engineering , Lanzhou University , Lanzhou , 730000 , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Information Science and Engineering , Lanzhou University , Lanzhou , 730000 , China","institution_ids":["https://openalex.org/I76214153"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5148230117","display_name":"Tat-SengChua","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tat-SengChua","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"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":9.999999747378752e-05,"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/referring-image-segmentation","display_name":"referring image segmentation","score":0.6460000276565552},{"id":"https://openalex.org/keywords/video-segmentation","display_name":"video segmentation","score":0.6579999923706055},{"id":"https://openalex.org/keywords/spatial-priors","display_name":"spatial priors","score":0.765999972820282},{"id":"https://openalex.org/keywords/knowledge-distillation","display_name":"knowledge distillation","score":0.8119999766349792},{"id":"https://openalex.org/keywords/dinov2","display_name":"DINOv2","score":0.9089999794960022},{"id":"https://openalex.org/keywords/occlusion","display_name":"occlusion","score":0.5529999732971191},{"id":"https://openalex.org/keywords/temporal-stability","display_name":"temporal stability","score":0.5180000066757202},{"id":"https://openalex.org/keywords/grounded-reasoning","display_name":"grounded reasoning","score":0.6669999957084656}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.717199981212616},{"id":"https://openalex.org/C177769412","wikidata":"https://www.wikidata.org/wiki/Q278090","display_name":"Prior probability","level":3,"score":0.6586999893188477},{"id":"https://openalex.org/C89600930","wikidata":"https://www.wikidata.org/wiki/Q1423946","display_name":"Segmentation","level":2,"score":0.6320000290870667},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.607200026512146},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5568000078201294},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.521399974822998},{"id":"https://openalex.org/C2778355321","wikidata":"https://www.wikidata.org/wiki/Q17079427","display_name":"Identity (music)","level":2,"score":0.4778999984264374},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.46149998903274536},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.4138999879360199},{"id":"https://openalex.org/C2779808786","wikidata":"https://www.wikidata.org/wiki/Q6664603","display_name":"Locality","level":2,"score":0.387800008058548},{"id":"https://openalex.org/C2911011789","wikidata":"https://www.wikidata.org/wiki/Q130741","display_name":"Hallucinating","level":2,"score":0.3847000002861023},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.37779998779296875},{"id":"https://openalex.org/C124504099","wikidata":"https://www.wikidata.org/wiki/Q56933","display_name":"Image segmentation","level":3,"score":0.37139999866485596},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.3589000105857849},{"id":"https://openalex.org/C2776436953","wikidata":"https://www.wikidata.org/wiki/Q5163215","display_name":"Consistency (knowledge bases)","level":2,"score":0.34200000762939453},{"id":"https://openalex.org/C22367795","wikidata":"https://www.wikidata.org/wiki/Q7625208","display_name":"Structured prediction","level":2,"score":0.3418999910354614},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.328900009393692},{"id":"https://openalex.org/C75553542","wikidata":"https://www.wikidata.org/wiki/Q178161","display_name":"A priori and a posteriori","level":2,"score":0.32760000228881836},{"id":"https://openalex.org/C2777508537","wikidata":"https://www.wikidata.org/wiki/Q7936620","display_name":"Visual reasoning","level":2,"score":0.3095000088214874},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.30410000681877136},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3012000024318695},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.2928999960422516},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.28949999809265137},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.28949999809265137},{"id":"https://openalex.org/C126042441","wikidata":"https://www.wikidata.org/wiki/Q1324888","display_name":"Frame (networking)","level":2,"score":0.2777000069618225},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.2694999873638153},{"id":"https://openalex.org/C146849305","wikidata":"https://www.wikidata.org/wiki/Q370766","display_name":"Ground truth","level":2,"score":0.2687999904155731},{"id":"https://openalex.org/C774472","wikidata":"https://www.wikidata.org/wiki/Q6760393","display_name":"Margin (machine learning)","level":2,"score":0.2590999901294708},{"id":"https://openalex.org/C2779010991","wikidata":"https://www.wikidata.org/wiki/Q2720909","display_name":"Artifact (error)","level":2,"score":0.25699999928474426},{"id":"https://openalex.org/C2777402240","wikidata":"https://www.wikidata.org/wiki/Q6783436","display_name":"Masking (illustration)","level":2,"score":0.2563000023365021},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.2547999918460846}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2608.24574","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.24574","pdf_url":"https://arxiv.org/pdf/2608.24574","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2608.24574","doi":"https://doi.org/10.48550/arxiv.2608.24574","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2608.24574","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:2608.24574","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.24574","pdf_url":"https://arxiv.org/pdf/2608.24574","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[{"id":"https://openalex.org/G1056251607","display_name":null,"funder_award_id":"62227807","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G6414949351","display_name":null,"funder_award_id":"U24B20186","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320313889","display_name":"Lanzhou University","ror":"https://ror.org/01mkqqe32"},{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7204267973.pdf","grobid_xml":"https://content.openalex.org/works/W7204267973.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7136316570","https://openalex.org/W6979324028","https://openalex.org/W4417086854","https://openalex.org/W7140221275","https://openalex.org/W4416247172","https://openalex.org/W4415798953","https://openalex.org/W7169871736","https://openalex.org/W7161276038","https://openalex.org/W4401200977","https://openalex.org/W7160527609"],"abstract_inverted_index":{"Video":[0,77],"multimodal":[1,199],"large":[2],"language":[3,6],"models":[4],"support":[5],"guided":[7],"video":[8,143,147,191],"segmentation,":[9],"but":[10],"they":[11],"often":[12],"show":[13],"spatio":[14],"temporal":[15,211],"inconsistencies,":[16],"e.g.,":[17],"jitter,":[18],"drift,":[19],"and":[20,58,124,134,151,167,175],"identity":[21,57],"switches.":[22],"These":[23,201],"failures":[24],"are":[25,30],"more":[26,84],"common":[27],"when":[28,34],"targets":[29],"partly":[31],"hidden":[32],"or":[33,197],"similar":[35],"objects":[36],"appear":[37],"nearby.One":[38],"likely":[39],"reason":[40],"is":[41,80,219],"that":[42,70,185,204],"current":[43],"training":[44],"lacks":[45],"explicit":[46],"spatial":[47,56,73,188,206],"priors,":[48],"which":[49],"makes":[50],"it":[51],"difficult":[52],"to":[53,82],"maintain":[54],"stable":[55,85],"shape":[59],"over":[60],"time.":[61],"We":[62],"present":[63],"PhysMLLMs,":[64],"a":[65,96,115,125],"training-stage":[66],"prior":[67,189,207],"injection":[68,208],"architecture":[69],"injects":[71],"physics-inspired":[72,205],"continuity":[74],"priors":[75],"into":[76],"MLLMs.":[78],"PhysMLLMs":[79,145,180],"designed":[81],"encourage":[83],"object-centered":[86],"representations":[87,113],"by":[88],"aligning":[89],"the":[90,186],"student":[91],"global":[92,111],"visual":[93,112],"representation":[94],"with":[95,154],"frozen":[97,116],"teacher":[98,118],"model":[99],"during":[100],"training.":[101],"Our":[102],"core":[103],"mechanism,":[104],"Global":[105],"Representation":[106],"Prior":[107],"Alignment":[108],"(REPA-Global),":[109],"distills":[110],"from":[114],"DINOv2":[117],"using":[119],"an":[120],"offline":[121],"embedding":[122],"cache":[123],"scheduled":[126],"distillation":[127],"plan.":[128],"This":[129],"design":[130],"keeps":[131],"inference":[132,138],"unchanged":[133],"does":[135],"not":[136],"add":[137],"time":[139],"cost.":[140],"Across":[141],"multiple":[142],"benchmarks,":[144,179],"improves":[146,190],"segmentation":[148,174],"mask":[149],"quality":[150],"cross-frame":[152],"consistency,":[153],"larger":[155],"gains":[156],"on":[157],"challenging":[158],"cases":[159],"involving":[160],"small":[161],"targets,":[162],"fast":[163],"motion,":[164],"occlusion,":[165],"distractors,":[166],"reasoning":[168],"queries.":[169],"On":[170],"single-frame":[171],"referring":[172],"image":[173],"representative":[176],"general":[177,198,215],"VLM":[178],"maintains":[181],"comparable":[182],"performance,":[183],"demonstrating":[184],"injected":[187],"consistency":[192],"without":[193],"compromising":[194],"image-level":[195],"grounding":[196],"capability.":[200,216],"results":[202],"suggest":[203],"can":[209],"improve":[210],"stability":[212],"while":[213],"preserving":[214],"The":[217],"code":[218],"available":[220],"at":[221],"https://github.com/tusu-code/20260121-icml2026-2.git.":[222]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-08-27T00:00:00"}