{"id":"https://openalex.org/W4415969880","doi":"https://doi.org/10.23919/icmu65253.2025.11219147","title":"Few-Shot Vision-Based Human Activity Recognition with MLLM-Based Visual Reinforcement Learning","display_name":"Few-Shot Vision-Based Human Activity Recognition with MLLM-Based Visual Reinforcement Learning","publication_year":2025,"publication_date":"2025-09-10","ids":{"openalex":"https://openalex.org/W4415969880","doi":"https://doi.org/10.23919/icmu65253.2025.11219147"},"language":null,"primary_location":{"id":"doi:10.23919/icmu65253.2025.11219147","is_oa":false,"landing_page_url":"https://doi.org/10.23919/icmu65253.2025.11219147","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 Fifteenth International Conference on Mobile Computing and Ubiquitous Networking (ICMU)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100568961","display_name":"Wenqi Zheng","orcid":null},"institutions":[{"id":"https://openalex.org/I135598925","display_name":"Kyushu University","ror":"https://ror.org/00p4k0j84","country_code":"JP","type":"education","lineage":["https://openalex.org/I135598925"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Wenqi Zheng","raw_affiliation_strings":["Kyushu University,Graduate School and Faculty of Information Science and Electrical Engineering,JAPAN"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Kyushu University,Graduate School and Faculty of Information Science and Electrical Engineering,JAPAN","institution_ids":["https://openalex.org/I135598925"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5084986446","display_name":"Yutaka Arakawa","orcid":"https://orcid.org/0000-0002-7156-9160"},"institutions":[{"id":"https://openalex.org/I135598925","display_name":"Kyushu University","ror":"https://ror.org/00p4k0j84","country_code":"JP","type":"education","lineage":["https://openalex.org/I135598925"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Yutaka Arakawa","raw_affiliation_strings":["Kyushu University,Graduate School and Faculty of Information Science and Electrical Engineering,JAPAN"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Kyushu University,Graduate School and Faculty of Information Science and Electrical Engineering,JAPAN","institution_ids":["https://openalex.org/I135598925"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I135598925"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"6"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10812","display_name":"Human Pose and Action Recognition","score":0.44850000739097595,"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/T10812","display_name":"Human Pose and Action Recognition","score":0.44850000739097595,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.3416999876499176,"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/T10444","display_name":"Context-Aware Activity Recognition Systems","score":0.05429999902844429,"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":0.7684999704360962},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.598800003528595},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5218999981880188},{"id":"https://openalex.org/keywords/activity-recognition","display_name":"Activity recognition","score":0.454800009727478},{"id":"https://openalex.org/keywords/reinforcement","display_name":"Reinforcement","score":0.4316999912261963},{"id":"https://openalex.org/keywords/domain","display_name":"Domain (mathematical analysis)","score":0.3686999976634979},{"id":"https://openalex.org/keywords/signature-recognition","display_name":"Signature recognition","score":0.3325999975204468}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.7684999704360962},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6182000041007996},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.598800003528595},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.59170001745224},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5218999981880188},{"id":"https://openalex.org/C121687571","wikidata":"https://www.wikidata.org/wiki/Q4677630","display_name":"Activity recognition","level":2,"score":0.454800009727478},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4390000104904175},{"id":"https://openalex.org/C67203356","wikidata":"https://www.wikidata.org/wiki/Q1321905","display_name":"Reinforcement","level":2,"score":0.4316999912261963},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.3686999976634979},{"id":"https://openalex.org/C74370796","wikidata":"https://www.wikidata.org/wiki/Q15924863","display_name":"Signature recognition","level":3,"score":0.3325999975204468},{"id":"https://openalex.org/C2779321571","wikidata":"https://www.wikidata.org/wiki/Q7936605","display_name":"Visual learning","level":2,"score":0.3239000141620636},{"id":"https://openalex.org/C77967617","wikidata":"https://www.wikidata.org/wiki/Q4677561","display_name":"Active learning (machine learning)","level":2,"score":0.32350000739097595},{"id":"https://openalex.org/C132900626","wikidata":"https://www.wikidata.org/wiki/Q7534733","display_name":"Sketch recognition","level":4,"score":0.28769999742507935},{"id":"https://openalex.org/C178253425","wikidata":"https://www.wikidata.org/wiki/Q162668","display_name":"Visual perception","level":3,"score":0.2671999931335449},{"id":"https://openalex.org/C2780660688","wikidata":"https://www.wikidata.org/wiki/Q25052564","display_name":"Multimodal learning","level":2,"score":0.2662000060081482},{"id":"https://openalex.org/C40506919","wikidata":"https://www.wikidata.org/wiki/Q7452469","display_name":"Sequence learning","level":2,"score":0.2583000063896179},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.25600001215934753}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.23919/icmu65253.2025.11219147","is_oa":false,"landing_page_url":"https://doi.org/10.23919/icmu65253.2025.11219147","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 Fifteenth International Conference on Mobile Computing and Ubiquitous Networking (ICMU)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":11,"referenced_works":["https://openalex.org/W1981781955","https://openalex.org/W2126579184","https://openalex.org/W2963795951","https://openalex.org/W4282981352","https://openalex.org/W4390692125","https://openalex.org/W4391128353","https://openalex.org/W4391342264","https://openalex.org/W4391342339","https://openalex.org/W4399146412","https://openalex.org/W4399426086","https://openalex.org/W4411403346"],"related_works":[],"abstract_inverted_index":{"Reinforcement":[0],"learning":[1,7,42,59,79,106],"in":[2,17,27,60,91],"Large":[3],"Reasoning":[4],"Models":[5],"enables":[6],"from":[8],"feedback":[9],"on":[10,68,180],"their":[11],"outputs,":[12],"making":[13],"it":[14],"particularly":[15],"valuable":[16],"scenarios":[18],"where":[19],"fine-tuning":[20],"data":[21],"is":[22,149],"limited.":[23],"However,":[24],"its":[25],"application":[26],"multi-modal":[28],"human":[29,45,99,127,170,182],"activity":[30,46,128,171,183],"recognition":[31,47,70,101,172,184],"(HAR)":[32],"domains":[33],"remains":[34],"largely":[35],"underexplored.":[36],"Our":[37],"work":[38],"extends":[39],"the":[40,44,56,61,64,76,82,88,92,126,146,153,163,191,194],"reinforcement":[41,58,78],"to":[43,120,169],"domain":[48],"with":[49,103,173],"multimodal":[50,115],"large":[51,116],"language":[52,117],"models.":[53],"By":[54],"incorporating":[55],"visual":[57,77],"training":[62],"process,":[63],"model's":[65],"generalization":[66],"ability":[67,85],"few-shot":[69],"can":[71,80,166],"be":[72,167],"greatly":[73],"improved.":[74],"Besides,":[75],"enhance":[81],"model":[83,118,148,165],"reasoning":[84,132],"and":[86,134,145,186],"enable":[87],"explainable":[89],"analysis":[90],"inference":[93],"stage.":[94],"We":[95],"name":[96],"our":[97,110],"Few-shot":[98],"Activity":[100],"method":[102],"Visual":[104],"reinfORcement":[105],"as":[107],"FAVOR.":[108],"Specifically,":[109],"approach":[111],"firstly":[112],"utilizes":[113],"a":[114,175],"(MLLM)":[119],"generate":[121],"multiple":[122],"candidate":[123],"responses":[124,138],"for":[125],"image,":[129],"each":[130],"containing":[131],"traces":[133],"final":[135],"answers.":[136],"These":[137],"are":[139],"then":[140],"evaluated":[141],"using":[142,152],"reward":[143],"functions":[144],"MLLM":[147,164],"subsequently":[150],"optimized":[151],"Group":[154],"Relative":[155],"Policy":[156],"Optimization":[157],"(GRPO)":[158],"algorithm.":[159],"In":[160],"this":[161],"way,":[162],"adapted":[168],"only":[174],"few":[176],"samples.":[177],"Extensive":[178],"experiments":[179],"4":[181],"datasets":[185],"5":[187],"different":[188],"settings":[189],"demonstrate":[190],"superiority":[192],"of":[193],"proposed":[195],"method.":[196]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-11-06T00:00:00"}