{"id":"https://openalex.org/W7140828582","doi":"https://doi.org/10.48550/arxiv.2603.23957","title":"PointRFT: Explicit Reinforcement Fine-tuning for Point Cloud Few-shot Learning","display_name":"PointRFT: Explicit Reinforcement Fine-tuning for Point Cloud Few-shot Learning","publication_year":2026,"publication_date":"2026-03-25","ids":{"openalex":"https://openalex.org/W7140828582","doi":"https://doi.org/10.48550/arxiv.2603.23957"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.23957","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.23957","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":"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"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2603.23957","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5130703441","display_name":"Yankai Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Yankai","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130689071","display_name":"Yiding Sun","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sun, Yiding","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101729503","display_name":"Qirui Wang","orcid":"https://orcid.org/0000-0003-3653-2984"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Qirui","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130644303","display_name":"Pengbo Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Pengbo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5077212970","display_name":"Chaoyi Lu","orcid":"https://orcid.org/0009-0007-9484-5838"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lu, Chaoyi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5130692741","display_name":"Dongxu Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Dongxu","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/T10719","display_name":"3D Shape Modeling and Analysis","score":0.39160001277923584,"subfield":{"id":"https://openalex.org/subfields/2206","display_name":"Computational Mechanics"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10719","display_name":"3D Shape Modeling and Analysis","score":0.39160001277923584,"subfield":{"id":"https://openalex.org/subfields/2206","display_name":"Computational Mechanics"},"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.1559000015258789,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.06210000067949295,"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/reinforcement-learning","display_name":"Reinforcement learning","score":0.7904999852180481},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.5511999726295471},{"id":"https://openalex.org/keywords/point-cloud","display_name":"Point cloud","score":0.5170999765396118},{"id":"https://openalex.org/keywords/cloud-computing","display_name":"Cloud computing","score":0.5011000037193298},{"id":"https://openalex.org/keywords/point","display_name":"Point (geometry)","score":0.49720001220703125},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.42289999127388},{"id":"https://openalex.org/keywords/foundation","display_name":"Foundation (evidence)","score":0.4009000062942505}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.7904999852180481},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7128000259399414},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5537999868392944},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.5511999726295471},{"id":"https://openalex.org/C131979681","wikidata":"https://www.wikidata.org/wiki/Q1899648","display_name":"Point cloud","level":2,"score":0.5170999765396118},{"id":"https://openalex.org/C79974875","wikidata":"https://www.wikidata.org/wiki/Q483639","display_name":"Cloud computing","level":2,"score":0.5011000037193298},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.49720001220703125},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4578000009059906},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.42289999127388},{"id":"https://openalex.org/C2780966255","wikidata":"https://www.wikidata.org/wiki/Q5474306","display_name":"Foundation (evidence)","level":2,"score":0.4009000062942505},{"id":"https://openalex.org/C26760741","wikidata":"https://www.wikidata.org/wiki/Q160402","display_name":"Perception","level":2,"score":0.3971000015735626},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.3668999969959259},{"id":"https://openalex.org/C88871306","wikidata":"https://www.wikidata.org/wiki/Q7208287","display_name":"Point process","level":2,"score":0.2957000136375427},{"id":"https://openalex.org/C59404180","wikidata":"https://www.wikidata.org/wiki/Q17013334","display_name":"Feature learning","level":2,"score":0.2800999879837036},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.27410000562667847},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.2736999988555908},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.2728999853134155},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.2662999927997589},{"id":"https://openalex.org/C67203356","wikidata":"https://www.wikidata.org/wiki/Q1321905","display_name":"Reinforcement","level":2,"score":0.25200000405311584}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.23957","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.23957","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2603.23957","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.23957","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":"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"},"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":{"Understanding":[0],"spatial":[1],"dynamics":[2],"and":[3,92,97,104],"semantics":[4],"in":[5,30,47],"point":[6,64,81,143],"cloud":[7,65,82,144],"is":[8,147],"fundamental":[9],"for":[10,80],"comprehensive":[11,109],"3D":[12,49,63,89],"comprehension.":[13],"While":[14],"reinforcement":[15,75],"learning":[16],"algorithms":[17],"such":[18],"as":[19],"Group":[20],"Relative":[21],"Policy":[22],"Optimization":[23],"(GRPO)":[24],"have":[25],"recently":[26],"achieved":[27],"remarkable":[28],"breakthroughs":[29],"large":[31],"language":[32],"models":[33,91,146],"by":[34],"incentivizing":[35],"reasoning":[36],"capabilities":[37],"through":[38],"strategic":[39],"reward":[40,96,99],"design,":[41],"their":[42],"potential":[43],"remains":[44],"largely":[45],"unexplored":[46],"the":[48,73,139],"perception":[50],"domain.":[51],"This":[52],"naturally":[53],"raises":[54],"a":[55,135],"pivotal":[56],"question:":[57],"Can":[58],"RL-based":[59],"methods":[60],"effectively":[61],"empower":[62],"fine-tuning?":[66],"In":[67],"this":[68],"paper,":[69],"we":[70,117],"propose":[71],"PointRFT,":[72],"first":[74],"fine-tuning":[76,125],"paradigm":[77],"tailored":[78],"specifically":[79],"representation":[83],"learning.":[84],"We":[85],"select":[86],"three":[87],"prevalent":[88],"foundation":[90,145],"devise":[93],"specialized":[94],"accuracy":[95],"dispersion":[98],"functions":[100],"to":[101],"stabilize":[102],"training":[103,115],"mitigate":[105],"distribution":[106],"shifts.":[107],"Through":[108],"few-shot":[110],"classification":[111],"experiments":[112],"comparing":[113],"distinct":[114],"paradigms,":[116],"demonstrate":[118],"that":[119],"PointRFT":[120],"consistently":[121],"outperforms":[122],"vanilla":[123],"supervised":[124],"(SFT)":[126],"across":[127],"diverse":[128],"benchmarks.":[129],"Furthermore,":[130],"when":[131],"organically":[132],"integrated":[133],"into":[134],"hybrid":[136],"Pretraining-SFT-RFT":[137],"paradigm,":[138],"representational":[140],"capacity":[141],"of":[142],"substantially":[148],"unleashed,":[149],"achieving":[150],"state-of-the-art":[151],"performance":[152],"particularly":[153],"under":[154],"data-scarce":[155],"scenarios.":[156]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-03-27T00:00:00"}