{"id":"https://openalex.org/W7171524726","doi":"https://doi.org/10.48550/arxiv.2607.23605","title":"Hybrid Advantage Estimation with Unified Critic for VLM Agentic Reinforcement Learning","display_name":"Hybrid Advantage Estimation with Unified Critic for VLM Agentic Reinforcement Learning","publication_year":2026,"publication_date":"2026-07-26","ids":{"openalex":"https://openalex.org/W7171524726","doi":"https://doi.org/10.48550/arxiv.2607.23605"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2607.23605","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.23605","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.2607.23605","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5143790598","display_name":"Wenxuan Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Wenxuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5143811749","display_name":"Yuhui Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Yuhui","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5002662014","display_name":"Donggang Jia","orcid":"https://orcid.org/0000-0002-1358-8718"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jia, Donggang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5020511969","display_name":"Xiaoqian Shen","orcid":"https://orcid.org/0000-0001-6284-520X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shen, Xiaoqian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5143858256","display_name":"Jian Ding","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ding, Jian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5022294742","display_name":"Ivan Viola","orcid":"https://orcid.org/0000-0003-4248-6574"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Viola, Ivan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128424967","display_name":"J\u00fcrgen Schmidhuber","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Schmidhuber, J\u00fcrgen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5143762303","display_name":"Mohamed Elhoseiny","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Elhoseiny, Mohamed","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":0.9620000123977661,"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":0.9620000123977661,"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":0.005799999926239252,"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/T12031","display_name":"Speech and dialogue systems","score":0.0032999999821186066,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.638700008392334},{"id":"https://openalex.org/keywords/unified-model","display_name":"Unified Model","score":0.5078999996185303},{"id":"https://openalex.org/keywords/factor","display_name":"Factor (programming language)","score":0.38690000772476196},{"id":"https://openalex.org/keywords/optimization-problem","display_name":"Optimization problem","score":0.33160001039505005},{"id":"https://openalex.org/keywords/constrained-optimization","display_name":"Constrained optimization","score":0.31790000200271606}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6776000261306763},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.638700008392334},{"id":"https://openalex.org/C45493050","wikidata":"https://www.wikidata.org/wiki/Q7884934","display_name":"Unified Model","level":2,"score":0.5078999996185303},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4745999872684479},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.3889000117778778},{"id":"https://openalex.org/C2781039887","wikidata":"https://www.wikidata.org/wiki/Q1391724","display_name":"Factor (programming language)","level":2,"score":0.38690000772476196},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3741999864578247},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.33160001039505005},{"id":"https://openalex.org/C55660270","wikidata":"https://www.wikidata.org/wiki/Q5164377","display_name":"Constrained optimization","level":2,"score":0.31790000200271606},{"id":"https://openalex.org/C136197465","wikidata":"https://www.wikidata.org/wiki/Q1729295","display_name":"Variety (cybernetics)","level":2,"score":0.28619998693466187},{"id":"https://openalex.org/C96250715","wikidata":"https://www.wikidata.org/wiki/Q965330","display_name":"Estimation","level":2,"score":0.26249998807907104}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2607.23605","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.23605","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.2607.23605","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.23605","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":[{"id":"https://metadata.un.org/sdg/16","display_name":"Peace, Justice and strong institutions","score":0.7609844207763672}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Large":[0],"Vision-Language":[1],"Models":[2],"(VLMs)":[3],"now":[4],"act":[5],"as":[6],"agents":[7],"in":[8,23],"interactive":[9],"environments,":[10,131],"where":[11,132],"success":[12,137],"requires":[13],"coherent":[14],"reasoning":[15],"and":[16,64,81,97,111,118,141,166],"decision-making":[17,30,130],"across":[18,127],"turns.":[19],"Although":[20],"end-to-end":[21],"training":[22],"agentic":[24],"environments":[25],"can":[26,91],"improve":[27],"such":[28],"multi-turn":[29,129],"abilities,":[31],"current":[32],"methods":[33],"mainly":[34],"rely":[35],"on":[36],"either":[37],"token-wise":[38],"optimization":[39,46,63],"over":[40,147],"concatenated":[41],"token":[42],"trajectories":[43],"or":[44],"turn-wise":[45,96],"with":[47,74,114],"uniform":[48],"within-turn":[49],"credit.":[50],"In":[51],"this":[52],"work,":[53],"we":[54,84,101,151],"establish":[55],"theoretical":[56],"formulations":[57],"for":[58,94,170],"the":[59,115,158,163],"two":[60],"levels":[61],"of":[62,78,125,139,145,162],"derive":[65],"a":[66,87,142],"hybrid":[67,116,164],"advantage":[68,117,165],"that":[69,86,107,157],"serves":[70],"both":[71,95],"objectives.":[72],"Furthermore,":[73,150],"an":[75,104,135,153],"appropriate":[76],"choice":[77],"discount":[79],"factor":[80],"learning":[82],"target,":[83],"prove":[85],"unified":[88,119],"critic":[89],"model":[90],"estimate":[92],"values":[93],"token-wise.":[98],"As":[99],"such,":[100],"propose":[102],"HyGAE,":[103],"actor-critic":[105],"framework":[106],"jointly":[108],"optimizes":[109],"token-":[110],"turn-level":[112],"objectives":[113],"critic.":[120],"We":[121],"conduct":[122],"extensive":[123],"evaluations":[124],"HyGAE":[126],"five":[128],"it":[133],"achieves":[134],"average":[136],"rate":[138],"91%":[140],"significant":[143],"improvement":[144],"10%":[146],"other":[148],"methods.":[149],"provide":[152],"in-depth":[154],"analysis":[155],"showing":[156],"exact":[159],"analytic":[160],"form":[161],"return":[167],"is":[168],"crucial":[169],"optimization.":[171],"Project":[172],"Page:":[173],"https://wx-zhang.github.io/hygae-web/.":[174]},"counts_by_year":[],"updated_date":"2026-07-29T09:40:50.615796","created_date":"2026-07-29T00:00:00"}