{"id":"https://openalex.org/W7163172424","doi":"https://doi.org/10.48550/arxiv.2606.01047","title":"Learning Multi-Modal Trajectory Policies for Data-Efficient Robotic Manipulation","display_name":"Learning Multi-Modal Trajectory Policies for Data-Efficient Robotic Manipulation","publication_year":2026,"publication_date":"2026-05-31","ids":{"openalex":"https://openalex.org/W7163172424","doi":"https://doi.org/10.48550/arxiv.2606.01047"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.01047","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.01047","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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.2606.01047","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5137667467","display_name":"Zijia Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Zijia","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5062441362","display_name":"Yuenan Hou","orcid":"https://orcid.org/0000-0002-2844-7416"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hou, Yuenan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5018850495","display_name":"Xinhua Jiang","orcid":"https://orcid.org/0000-0002-1379-7456"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiang, Xinhua","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137674780","display_name":"Yu Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Yu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137709650","display_name":"Weijie Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Weijie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5137661312","display_name":"Li Liu","orcid":"https://orcid.org/0000-0002-8655-2644"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Li","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":0.28940001130104065,"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":0.28940001130104065,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.1639000028371811,"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"}},{"id":"https://openalex.org/T10653","display_name":"Robot Manipulation and Learning","score":0.13019999861717224,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/trajectory","display_name":"Trajectory","score":0.7301999926567078},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.5662999749183655},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.5619999766349792},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.5378000140190125},{"id":"https://openalex.org/keywords/routing","display_name":"Routing (electronic design automation)","score":0.49480000138282776},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.4546999931335449},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.4431999921798706},{"id":"https://openalex.org/keywords/structured-prediction","display_name":"Structured prediction","score":0.43130001425743103},{"id":"https://openalex.org/keywords/router","display_name":"Router","score":0.4124999940395355}],"concepts":[{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.7301999926567078},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6977999806404114},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5662999749183655},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.5619999766349792},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.5378000140190125},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4961000084877014},{"id":"https://openalex.org/C74172769","wikidata":"https://www.wikidata.org/wiki/Q1446839","display_name":"Routing (electronic design automation)","level":2,"score":0.49480000138282776},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4683000147342682},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.4546999931335449},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.4431999921798706},{"id":"https://openalex.org/C22367795","wikidata":"https://www.wikidata.org/wiki/Q7625208","display_name":"Structured prediction","level":2,"score":0.43130001425743103},{"id":"https://openalex.org/C2775896111","wikidata":"https://www.wikidata.org/wiki/Q642560","display_name":"Router","level":2,"score":0.4124999940395355},{"id":"https://openalex.org/C2780226545","wikidata":"https://www.wikidata.org/wiki/Q6888030","display_name":"Modality (human\u2013computer interaction)","level":2,"score":0.3869999945163727},{"id":"https://openalex.org/C90509273","wikidata":"https://www.wikidata.org/wiki/Q11012","display_name":"Robot","level":2,"score":0.38499999046325684},{"id":"https://openalex.org/C34413123","wikidata":"https://www.wikidata.org/wiki/Q170978","display_name":"Robotics","level":3,"score":0.36739999055862427},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.36070001125335693},{"id":"https://openalex.org/C2780735816","wikidata":"https://www.wikidata.org/wiki/Q28324931","display_name":"Incremental learning","level":2,"score":0.3601999878883362},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.3416999876499176},{"id":"https://openalex.org/C173246807","wikidata":"https://www.wikidata.org/wiki/Q7833062","display_name":"Trajectory optimization","level":3,"score":0.3384999930858612},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.32350000739097595},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.3156000077724457},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.31529998779296875},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.3089999854564667},{"id":"https://openalex.org/C58328972","wikidata":"https://www.wikidata.org/wiki/Q184609","display_name":"Expert system","level":2,"score":0.2822999954223633},{"id":"https://openalex.org/C59404180","wikidata":"https://www.wikidata.org/wiki/Q17013334","display_name":"Feature learning","level":2,"score":0.2563000023365021},{"id":"https://openalex.org/C185429906","wikidata":"https://www.wikidata.org/wiki/Q1130160","display_name":"Estimator","level":2,"score":0.2554999887943268},{"id":"https://openalex.org/C81074085","wikidata":"https://www.wikidata.org/wiki/Q366872","display_name":"Motion planning","level":3,"score":0.25189998745918274}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.01047","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.01047","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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.2606.01047","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.01047","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Robotic":[0],"manipulation":[1],"requires":[2],"the":[3,145,169,180,193],"effective":[4],"integration":[5],"of":[6,196],"heterogeneous":[7,27,120],"inputs,":[8],"including":[9],"visual":[10],"observations,":[11],"language":[12],"instructions,":[13],"and":[14,40,74,107,115,127,135],"trajectory":[15,88],"representations,":[16],"to":[17,37,64,101,131],"generate":[18],"accurate":[19],"actions.":[20],"Existing":[21],"transformer-based":[22],"policies":[23],"typically":[24],"process":[25],"these":[26],"modalities":[28],"within":[29],"a":[30,52,86,97,109,161],"shared":[31],"parameter":[32],"space,":[33],"which":[34],"often":[35,62],"leads":[36],"modality":[38],"interference":[39],"inefficient":[41],"representation":[42,67],"learning,":[43],"especially":[44],"in":[45,70,164],"data-scarce":[46],"scenarios.":[47],"While":[48],"Mixture-of-Experts":[49],"(MoE)":[50],"offers":[51],"scalable":[53],"solution":[54],"through":[55],"expert":[56,72,75,117,133],"specialization,":[57],"conventional":[58],"routing":[59,126,138],"mechanisms":[60],"are":[61],"sensitive":[63],"such":[65],"cross-modal":[66,110],"discrepancies,":[68],"resulting":[69],"unstable":[71],"assignment":[73,118],"collapse.":[76],"In":[77],"this":[78],"work,":[79],"we":[80,95],"propose":[81],"MATE":[82,151],"(Multi-ModAl":[83],"TrajEctory":[84],"Policies),":[85],"novel":[87],"prediction":[89],"framework":[90],"built":[91],"upon":[92],"MoE.":[93],"Specifically,":[94],"introduce":[96],"Multi-Modal":[98],"MoE":[99],"architecture":[100],"achieve":[102],"fine-grained":[103],"sub-token":[104],"feature":[105],"decoupling,":[106],"design":[108],"cosine":[111],"router":[112],"for":[113,187],"stable":[114],"scale-invariant":[116],"across":[119],"modalities.":[121],"We":[122],"further":[123,191],"employ":[124],"temperature-controlled":[125],"stochastic":[128],"noise":[129],"injection":[130],"improve":[132],"balance":[134],"prevent":[136],"premature":[137],"collapse":[139],"under":[140,156],"scarce":[141],"demonstrations.":[142],"Experiments":[143],"on":[144,174],"LIBERO":[146],"benchmark":[147],"show":[148],"that":[149,179],"our":[150,197],"consistently":[152],"outperforms":[153],"prior":[154],"work":[155],"data":[157],"scarcity.":[158],"It":[159],"achieves":[160],"4.75%":[162],"improvement":[163],"average":[165],"success":[166],"rate":[167],"over":[168],"trajectory-guided":[170],"counterpart.":[171],"Real-world":[172],"experiments":[173],"robotic":[175,189],"ping-pong":[176],"also":[177],"suggest":[178],"predicted":[181],"trajectories":[182],"can":[183],"provide":[184],"useful":[185],"guidance":[186],"downstream":[188],"execution,":[190],"indicating":[192],"practical":[194],"feasibility":[195],"algorithm.":[198]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-03T00:00:00"}