{"id":"https://openalex.org/W7161962289","doi":"https://doi.org/10.48550/arxiv.2605.21414","title":"PointACT: Vision-Language-Action Models with Multi-Scale Point-Action Interaction","display_name":"PointACT: Vision-Language-Action Models with Multi-Scale Point-Action Interaction","publication_year":2026,"publication_date":"2026-05-20","ids":{"openalex":"https://openalex.org/W7161962289","doi":"https://doi.org/10.48550/arxiv.2605.21414"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.21414","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.21414","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,"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"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.21414","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5136694013","display_name":"Shizhe Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Shizhe","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5092693576","display_name":"Paul Pacaud","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pacaud, Paul","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5136693878","display_name":"Cordelia Schmid","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Schmid, Cordelia","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"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":0.6625999808311462,"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":0.6625999808311462,"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":0.181099995970726,"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/T10709","display_name":"Social Robot Interaction and HRI","score":0.01679999940097332,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/point-cloud","display_name":"Point cloud","score":0.6018000245094299},{"id":"https://openalex.org/keywords/bottleneck","display_name":"Bottleneck","score":0.5333999991416931},{"id":"https://openalex.org/keywords/action","display_name":"Action (physics)","score":0.508400022983551},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.4821999967098236},{"id":"https://openalex.org/keywords/point","display_name":"Point (geometry)","score":0.45680001378059387},{"id":"https://openalex.org/keywords/robot","display_name":"Robot","score":0.4113999903202057},{"id":"https://openalex.org/keywords/decoding-methods","display_name":"Decoding methods","score":0.3978999853134155}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7200000286102295},{"id":"https://openalex.org/C131979681","wikidata":"https://www.wikidata.org/wiki/Q1899648","display_name":"Point cloud","level":2,"score":0.6018000245094299},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5544000267982483},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.5333999991416931},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.508400022983551},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.4821999967098236},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.45680001378059387},{"id":"https://openalex.org/C90509273","wikidata":"https://www.wikidata.org/wiki/Q11012","display_name":"Robot","level":2,"score":0.4113999903202057},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.3978999853134155},{"id":"https://openalex.org/C151201525","wikidata":"https://www.wikidata.org/wiki/Q177239","display_name":"Limit (mathematics)","level":2,"score":0.35989999771118164},{"id":"https://openalex.org/C131584629","wikidata":"https://www.wikidata.org/wiki/Q4308705","display_name":"Coupling (piping)","level":2,"score":0.329800009727478},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.31949999928474426},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.3165000081062317},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.30390000343322754},{"id":"https://openalex.org/C2778751112","wikidata":"https://www.wikidata.org/wiki/Q835016","display_name":"Window (computing)","level":2,"score":0.27959999442100525},{"id":"https://openalex.org/C89611455","wikidata":"https://www.wikidata.org/wiki/Q6804646","display_name":"Mechanism (biology)","level":2,"score":0.2768999934196472},{"id":"https://openalex.org/C2780806968","wikidata":"https://www.wikidata.org/wiki/Q6045196","display_name":"Interaction model","level":2,"score":0.27619999647140503},{"id":"https://openalex.org/C2777894999","wikidata":"https://www.wikidata.org/wiki/Q4781758","display_name":"Approx","level":2,"score":0.25440001487731934}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.21414","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.21414","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,"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":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.21414","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.21414","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,"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"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Vision-Language-Action":[0],"(VLA)":[1],"models":[2],"have":[3],"shown":[4],"strong":[5],"potential":[6],"for":[7,44,186,203],"general-purpose":[8],"robotic":[9],"manipulation":[10,48],"by":[11,140],"leveraging":[12],"large":[13],"pretrained":[14,149,180,200],"vision-language":[15,157],"backbones.":[16],"However,":[17],"most":[18],"existing":[19],"VLAs":[20],"rely":[21],"primarily":[22],"on":[23,107,142],"2D":[24,181],"visual":[25],"representations,":[26],"which":[27],"limit":[28],"their":[29],"ability":[30],"to":[31,92,95],"reason":[32],"about":[33],"fine-grained":[34],"geometry":[35,178],"and":[36,46,100,110,113,119,161,188],"spatial":[37],"grounding":[38],"-":[39],"capabilities":[40],"that":[41,63,173],"are":[42],"essential":[43],"precise":[45],"robust":[47,187],"in":[49],"3D":[50,66,177,201],"environments.":[51],"In":[52],"this":[53],"paper,":[54],"we":[55],"propose":[56],"PointACT,":[57],"a":[58,78],"dual-system":[59,120],"3D-aware":[60,204],"VLA":[61,121,205],"policy":[62],"integrates":[64],"hierarchical":[65,176],"point":[67,127],"cloud":[68,128],"representations":[69,183,202],"directly":[70],"into":[71],"the":[72,108,143,156,162,197],"action":[73,90,163],"decoding":[74],"process.":[75],"PointACT":[76,106,130],"employs":[77],"multi-scale":[79],"point-action":[80],"interaction":[81],"mechanism":[82],"with":[83,126,151,179],"efficient":[84],"bottleneck":[85],"window":[86],"self-attention,":[87],"enabling":[88],"evolving":[89],"tokens":[91],"densely":[93],"attend":[94],"both":[96,135],"local":[97],"geometric":[98],"detail":[99],"global":[101],"scene":[102],"structure.":[103],"We":[104],"evaluate":[105],"LIBERO":[109],"RLBench":[111],"benchmarks":[112],"systematically":[114],"compare":[115],"it":[116],"against":[117],"monolithic":[118],"baselines,":[122],"including":[123],"variants":[124],"augmented":[125],"inputs.":[129],"achieves":[131],"consistent":[132],"improvements":[133],"across":[134],"benchmarks,":[136],"increasing":[137],"success":[138],"rates":[139],"10%":[141],"challenging":[144],"RLBench-10Tasks":[145],"suite":[146],"over":[147],"state-of-the-art":[148],"VLAs,":[150],"even":[152],"larger":[153],"gains":[154],"when":[155],"backbone":[158],"is":[159,165,184],"frozen":[160],"expert":[164],"trained":[166],"from":[167],"scratch.":[168],"Extensive":[169],"ablation":[170],"studies":[171],"demonstrate":[172],"tightly":[174],"coupling":[175],"semantic":[182],"critical":[185],"spatially":[189],"grounded":[190],"robot":[191],"control.":[192],"Our":[193],"results":[194],"also":[195],"highlight":[196],"promise":[198],"of":[199],"policies.":[206]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-05-22T00:00:00"}